TACTICAL MULTISENSOR RECONNAISSANCE (U) VOL. 4 TECHNOLOGY SURVEY

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP67B00657R000300210001-4
• Release Decision: RIFPUB
• Original Classification: S
• Document Page Count: 201
• Document Creation Date: December 22, 2016
• Document Release Date: July 8, 2009
• Sequence Number: 1
• Publication Date: June 15, 1965
• Content Type: REPORT

Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
]' 8452-65-42
SECRET
SPECIAL HANDLING
TACTICAL MULTISENSOR
RECONNAISSANCE (U)
VOL. 4
TECHNOLOGY SURVEY
15 JUNE , 1965
SPECIAL HANDLING
SECRET
TM-65- 4
COPY NO._
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
NW
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
2. Photographic Sensors . . . . . . . . . . . . . . . . . . . . . . 2-1
2.1 General Camera Characteristics . . . . . . . . . . . . . . 2-2
2.2 Frame Cameras versus Other Types ofCameras . . . . . . . . . . . 2-10
2.3 Strip Cameras . . . . . . . . . . . . . . . . . . . . . . . . . . 2-10
2.4 Panoramic Cameras . . . . . . . . . . . . . . . . . . . . . . . . 2-13
2.5 Representative Cameras . . . . . . . . . . . . . . . . . . . . . 2-17
2.6 Stabilized Mounts . . . . . . . . . . . . . . . . . . . . . . 2-17
2.7 Night Photography . . . . . . . . . . . . . . . . . . . . . . . . 2-20
2.8 Camera Control Systems . . . . . . . . . . . . . . . . . . . . . 2-23
3. Photographic Data Processing . . . . . . . . . . . . . . . . . . . . . 3-1
3.1 Inflight Processing . . . . . . . . . . . . . . . . . . . . . . . 3-1
3.2 Ground Processing . . . . . . . . . . . . . . . . . . . . . . . . 3-9
3.3 Rapid Processing . . . . . . . . . . . . . . . . . . . . . . . . 3-16
3.4 Printing . . . . . . . . . . . . . . . . . . . . . . 3-18
3.5 Automatic Contrast Control (Dodging) . . . . . . . . . . . . . . 3-23
3.6 Duplication Process . . . . . . . . . . . . . . . . . . . . . . . 3-31
4. Photographic Data Reduction . . . . . . . . . . . . . . . . . . . . . 4-1
4.1 Viewers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1
4.2 Reference Material . . . . . . . . . . . . . . . . . . . . . . . 4-6
4.3 Other Coverage . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
4.4 Data Block . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7
4.5 Photographic Records . . . . . . . . . . . . . . . . . . . . . . 4-7
4.6 Data Reduction Flow . . . . . . . . . . . . . . . . . . . . . . . 4-9
5. Photographic Output Products . . . . . . . . . . . . . . . . . . . . . 5-1
5.1 Photo Processing . . . . . . . . . . . . . . . . . . . . . . . . 5-1
5.2 Photo Duplication . . . . . . . . . . . . . . . . . . . . . . . . 5-3
5.3 Production Volume. . . . . . . . . . . . . . . . . . . . . . 5-7
5.4 Photo Products . . . . . . . . ... . . . . . . . . . . . . . . . 5-8
6. High Resolution Side Looking Radar, , , , , , , , , , , , , , , , , , 6-1
6.1 Synthetic Aperture Radar 6-5
6.2 Moving Target Indication (MTI) 6-36
6.3 Cross-Section Threshold Indicator, , , , , 6-45
6.4 Example Parameters , , , , , , , , , , , , , , , , , , , , , , , 6-45
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
F , z :r
SPECIAL HANDLING
CONTENTS (Continued)
7. Infrared. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1
7.1 Scene Characteristics . . . . . . . . . . . . . . . . . . . . . . 7-3
7.2 Detection Methods . . . . . . . . . . . . . . . . . . . . . . . . 7-4
7.3 Functional Elements . . . . . . . . . . . . . . . . . . . . . 7-7
7.4 Infrared Reconnaissance System Development . . . . . . . . . . . 7-10
7.5 Infrared Parametric Analysis . . . . . . . ? . . . . 7-12
7.6 Design Parameter Summary Illustrating Two Multi-Channel Systems. 7-17
8. Tactical ELINT . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
8.1 Emitter Characteristics . . . . . . . . . . . . . . . . . . . . . 8-1
8.2 Basic ELINT Functions . . . . . . . . . . . . . . . . . . . . . . 8-5
8.3 Basic System Components . . . . . . . . . . . . . . . . . . . . . 8-5
8.4 Data Flow Reduction. . . . . . . . . . . . . . . . . . . . . . . 8-6
8.5 Typical System . . . . . . . . . . . . . . . . . . . . . . . . . 8-7
iv SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w N
SPECIAL HANDLING
2-1. Lens-Film Resolution Required for Basic Ground Resolution. . . . . 2-4
2-2. Angular Coverage of Different Focal Length Lenses on Standard Film
Formats . . . . . . . . . . . . . . . . . . . . . . . . . . 2-6
2-3. Equivalent Ground Distance Covered (in feet) by a Vertical Camera
of Given Angular Coverage from Seven Representative Altitudes. . 2-6
2-4. Cycle Times Required at 55 percent Overlap for Various Angular
Coverages, at Representative Altitudes and Ground Speeds . . . . . 2-8
2-5. Representative Listing of Frame, Strip and Panoramic Cameras . . . 2-18
2-6. Characteristics of M-112, M-123 and XM-16 (Daisy) Cartridges . . . 2-22
3-1. Processing Equipment . . . . . . . . . . . . . . . . . . . . . 3-19
3-2. Granularity Comparison for Several Films . . . . . . . . . . . . . 3-44
3-3. Resolving Power Determinations on EK Type 4404 Film for Different
Chemical Formulations . . . . . . . . . . . . . . . . . . . . . . . 3-44
3-4. Granularity Comparison . . . . . . . . . . . . . . . . . . 3-52
3-5. Resolving Power Comparison (c/mm) . . . . . . . . . . . . . . . . . 3-52
4-1. Photographic Data Reduction Viewers . . . . . . . . . . . . . . . . 4-2
5-1. Operating Characteristics of Continuous Printers . . . . . . . . . 5-6
6-1. Target Radar Cross-Sections . . . . . . . . . . . . . . . . . . . . 6-44
6-2. Example of SLR Parameters . . . . . . . . . . . . . . . . . . . . . 6-46
7-1. Thermal Detection Regions . . . . . . . . . . . . . . . . . . . . . 7-5
7-2. Summary of Design Parameters, Design No. 1 . . . . . . . . . . . . 7-18
7-3. Summary of Design Parameters, Design No. 2 . . . . . . . . . . 7-19
7-4. Temperature Contrast for Signal/Noise of 7, Design No. 1 . . . . . 7-21
7-5. Temperature Contrast for Signal/Noise of 7, Design No. 2 . . . . 7-21
8-1. Functional Classification of Emitters . . . . . . . . . . . . . . . 8-2
8-2. Frequency Coverage for ELINT Detectors . . . . . . . . . . . . . . 8-8
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
SPECIAL HA-NDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
2-1.
Strip Camera Schematic . . . . . . . . . . . . . . . . . . . . . .
2-12
2-2.
HYAC-Type Panoramic Camera . . . . . . . . . . . . . . . . . . . .
2-14
2-3.
Direction Scanning Camera with 360 Degree Coverage . . . . . . .
2-15
2-4.
Ground Coverage Obtained with (a) Frame, (b) Strip, and (c)
Panoramic Cameras o . . . . . . . . . . . . . . . . . . . . . .
2-16
2-5.
Interrelationship of Terrain Clearance to Light Energy . . . . . .
2-25
2-6.
Speed Heights Cycling Rate Interrelationship . . . . . . . . . . .
2-26
3-1.
Change in Density for Fine Constant Exposure Values with Changes
in Agitation . . . . . . . . . . . . . . . . . . . . . . . . .
3-12
3-2.
Photographic Laboratory ES-14 or FM-25 . . . . . . . . . . . . . .
3-14
3-3.
Effect of Sample Atmospheric Luminance on Log Camera Exposure. . .
3-35
3-4.
Typical Tone Reproduction Cycle in High Altitude Aerial
Photography (Moderate Haze Condition), Showing Perfect and Actual
Reproduction of Original Scene . . . . . . . . . . . . . . . . . .
3-36
3-5.
Typical Tone Reproduction Cycle in High Altitude Aerial
Photography (Extremely Clear Weather), Showing Perfect and Actual
Reproduction of Original Scene . . . . . . . . . . . . . . . . . .
3-38
3-6.
Typical Reproduction Cycle for High Altitude Aerial Photography
Showing Actual Reproduction for Sample Case, and Desired
Characteristics of Positive Duping Film for Perfect Reproduction .
3-39
3-7.
Desired Characteristic Curves of Positive Duping Film for Perfect
Reproduction of Ground Scene Tonal Relationships for Three Haze
Conditions and 4404 Negative Materials as Shown in Fig. 6-7. . . .
3-40
3-8.
Modulation Transfer Function of Plus-X Aerographic (5401) in
Developers D-19 and D-76 (EK Data) . . . . . . . . . . . . . . .
3-43
3-9.
Characteristic Curves for EK Type 4404 Emulsion, Showing Increase
in Film Speed Possible with High Energy Developer MX-577 (Special
EK Developer) as Compared with Normal D-19 Processing . . . . . .
3-46
3-10.
l/E Speed versus Fog, S0-243 . . . . . . . . . . . . . . . . . . .
3-47
3-11.
272.5X Photomicrographs, Comparing Effects of Point Source with
Diffuse Illumination When Printing from Negative Material which
Contains a Layer of Quartz Crystals . . . . . . . . . . . . . . . .
3-55
3-12.
RMS Granularity Comparison of LIBR versus Conventional Processing
of 8430 Film . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-56
3-13.
Modulation Transfer Function Comparison of LIBR versus
Conventional Processing (Developed for 6 minutes at 68?F in D-19)
of 8430 Film; MTF Curves Have Had the Transfer Function of the
Microdensitometer Removed . . . . . . . . . . . . . . . . . . . .
3-58
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w- I I.
SPECIAL HANDLING
FIGURES (Continued)
5-1. Film Processing Production Chart . . . . . . . . . . . . . . . . . 5-2
5-2. Time to First Access . . . . . . . . . . . . . . . . . . . . . 5-4
6-1. Functional Elements for Radar System . . . . . . . . . . . . . . . 6-6
6-2. Azimuth Resolution for Three Cases: (a) Conventional, (b)
Unfocused, (c) Focused . . , . . . , 6-10
6-3. Geometry for Unfocused Synthetic Antenna 6-10
6-4. Side Looking Radar Pattern from Aircraft . . . . . . . . . . . . . 6-14
6-5. Calculation of R . . . . . . . . . . . . . . . . . . . . . . . . . 6-19
6-6. Recorder Configuration . . . . . . . . . . . . . . . . . . . . . . 6-26
6-7. Schematic Diagram of a Synthetic Aperture Radar, Optical Signal
Processor. . . . . . . . . . . . . . . . . . . . . . . . . . . 6-32
6-8. Appearance of Signal History . . . . . . . . . . . . . . . . . . . 6-33
6-9. Appearance of Signal History . . . . . . . . . . . . . . . . . . . 6-34
6-10. Photograph Made Using Synthetic Aperture Radar . . . . . . . . . . 6-37
6-11. Photograph Made Using Synthetic Aperture Radar . . . . . . . . . . 6-38
6-12. Photograph Made Using Synthetic Aperture Radar . . . . . . . . . . 6-39
6-13. Azimuth Distance Placement Error versus Range (MTI Operation
Velocity Uncertainty = ?30 fps). . . . . . . . . . . . . . . . . 6-42
6-14. Signal Cancellation for Slow Moving Targets. . . . . . . . . . . . 6-43
7-1. Functional Elements for Infrared System . . . . . . . . . . . . . . 7-6
7-2. Fraction of the Total Field versus Linear Size for Several
Altitudes (0? Nadir Angle) . . . . . . . . . . . . . . . . . . . . 7-22
7-3. Fraction of the Total Field versus Linear Size for Several
Altitudes (45? Nadir Angle) . . . . . . . . . . . . . . . . . . . . 7-23
8-1. Soviet Radar Spectrum Occupancy. . . . . . . . . . . . . . . . . 8-3
8-2. Typical System . . . . . . . . . . . . . . . . . . . . . . . . . . 8-9
viii SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
The technology surrounding multisensor reconnaissance operations requires a
breadth of understanding beyond the general knowledge of a specific sensor tech-
nology. This survey brings together the basic knowledge of principles and tech-
niques surrounding each of the four basic sensors commonly used for "multisensor"
reconnaissance. There sensors encompass photography, infrared, side looking
radar, and electronic intelligence. The information from each sensor technology
is designed to provide a basic understanding of the building blocks with which
each must inherently deal.
The contents of this document are by their nature a combination of basic
and transient data. In that light, further modification of the information
should be expected from time to time.
This volume is the technology survey for the Tactical Multisensor Recon-
naissance document. It contains the detail information which has been purposely
abbreviated in the preceding volumes in order that the reader not become mired
in detail not pertinent to the text.
SPECIAL HANDLING 1-1
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
NW %W
SPECIAL HANDLING
Photographic data acquisition in a tactical situation is a continuing
process. Prior to initial data acquisition, maps or course charts are usually
available, augmented by reports of visual observation and other intelligence
data. The commanders involved usually have a reasonable idea of what to look
for and where continuing activity should be monitored. However, data acquisi-
tion requirements change depending upon the operation that is involved.
Ground forces need various types of information. Artillery commanders
want low distortion in their photography so that range and azimuth to potential
targets can readily be determined. Infantry commanders want maximum detail, but
over relatively small areas. Armored, units want stereo coverage to enable them
to scale heights of terrain and obstacles and measure grades.
Air forces want confirmation of strike effectiveness; they are generally
more interested in the overall picture: road and rail networks, towns and villages,
location of airfields and possible emergency landing sites.
Operations and Intelligence want to see repeated coverage to detect changes
since their last "look". Higher Headquarters want large area coverage and gen-
eral surveillance type of information.
Whatever the requirement, all information is wanted quickly, plus being
easily legible, tilted, annotated and, if possible, fully interpreted. In addi-
tion, operations cannot be limited to daytime only, and provision must be made
for night-time coverage which requires artificial illumination.
Unfortunately, different requirements call for contradictory solutions.
Thus, the long focal length, high-altitude spotting camera, ideal for bomb
assessment work, is unsuitable for obtaining close detailed information for
infantry support. Therefore, in response to many different needs, many different
SPECIAL. HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
JIM?
SPECIAL HANDLING
types of cameras have become available over the years. Some of these were
designed for specific applications, while others represent compromises to make
them more versatile and generally acceptable. We will now consider some of the
characteristics which determine the use of a camera. Initial discussion will
be primarily concerned with the frame camera; the strip and the panoramic camera
will be covered further along in the text.
There is a basic division between a high-altitude application in which a
relatively large area can be covered with a reasonable amount of group detail,
and a low-altitude application aimed at obtaining the finest possible ground
detail over a relatively restricted area. This leads directly to considera-
tions of:
1. Focal Length
2. Ground Resolution
3. Angular Coverage
4. Camera Size
2.1.1 Focal Length Versus Ground Resolution
Since the scale of the resultant photography is equal to the ratio of focal
length over altitude, the focal length primarily determines how small an object
can be resolved from a given altitude. Of course, this is an over-simplifica-
tion, since many other factors will affect the final ground resolution. (It is
in accord, however, with the "first law of aerial reconnaissance" formulated by
General George Goddard: "There is no substitute for Focal Length.") The choice
of focal length therefore depends on how fine an object the user needs to re-
solve on the ground and from what altitude.
Typically, in the arbitrary altitude ranges of 300 to 1,500 feet for a low
altitude range and above 30,000 feet for a high altitude range, the following
resolutions would be expected:
Low Altitude: 4 inches to 1-foot resolution
High Altitude: I to 5-foot resolution
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
With the lens-film resolutions currently available, focal length selections for
the different altitudes can be made. Using presently available equipment, and
Using presently available equipment, and varying with the film type and
format, the obtainable lower and upper resolution boundaries can be arbitrarily
set as follows:
15 to 20 lines per millimeter can readily be obtained in nearly all cases
100 lines per millimeter or more is difficult to obtain operationally and
require rather sophisticated equipment.
Itek and other companies have produced equipment that exceeds the 100 lines per
millimeter upper boundary consistently; some systems do not always meet the 15
to 20 lines per millimeter taken as the lower boundary. These represent, how-
ever, generally valid limitations.
A table can now be constructed which shows the resolution (in lines per
millimter) for various focal lengths required to meet the desired ground resolu-
tions from representative altitudes. Table 2-1 shows these values for focal
lengths from 1.5 to 24 inches.
A number of conclusions can be drawn from this table.
1. 1.5 to 2-inch focal lengths will comfortably accommodate requirements
to 1,000 feet altitude.
2. 2 to 3-inch focal lengths will handle requirements to altitudes of 2000
feet.
3. 12-inch focal lengths will be adequate at the lower end of the high-
altitude range, and will be satisfactory at the upper end of this range for
ground resolutions beyond two feet.
4. If a single focal length had to be chosen to cover both high and low
altitudes, a 6-inch focal length would be most satisfactory - at least for the
low and medium speed aircraft.
2.1.2 Angular Coverage
The angular coverage obtained with lenses of a given focal length varies
with the chosen film format. The cross-track angular coverage determines how
wide a swath of terrain can be recorded from a given altitude, while the
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Table 2-1 - Lens-Film Resolution Required for Desired
Ground Resolution
Ground
Focal
Lens-Film Resolution (1/m) at
Resolution
Length
1000 ft
2000 ft 30,000 ft 50,000 ft
4 in
1.5 in
80
160
1 ft
27
53
4 in
2 in
60
120
1 ft
20
40
4 in
3 in
40
80
1 ft
13
26
4 in
6 in
20
40
1 ft
7
13
197
3 ft
67
110
4 in
12 in
197
1 f t
98
165
3 ft
33
55
4 in
24 in
150
250
1 ft
50
82
2-4 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
11W %0
SPECIAL HANDLING
fore-and-aft, or along-track, angular coverage determines, for a given aircraft
velocity and altitude, the rate at which photographs must be taken to obtain
continuous coverage of the terrain. Using conventional frame cameras, and for
a particular film format, a shorter focal length provides an increase in the
width of the swath of terrain that can be covered and a decrease in the cycling
time (i.e., frame rate).
The trade-offs in this case are that the shorter the lens focal length, the
smaller will be the scale of the resultant photography from any given altitude,
and, in general, the coarser will be the ground resolution. Moreover, it is
more difficult to achieve high resolution in lenses of wide angular coverage
than in lenses with a relatively narrow field of view. This also applies to
making lenses of wide relative aperture, which results in fast exposure times.
It is very difficult to make "fast" wide-angle lenses of reasonable quality.
Table 2-2 shows the angular coverage in degrees of different focal length
lenses on the three most commonly used formats - 70 mm, 5-inch and 9 1/2-inch
film.
Table 2-3 has been calculated for a number of representative altitudes to
give the reader a feeling for the equivalent ground distance covered by a
vertical camera with the angular coverages calculated in Table 2-1.
Table 2-3 illustrates one of the problems of present day tactical recon-
naissance. As aircraft fly lower and lower, each photograph covers an even
smaller area. As an example, an angular coverage of 41.1 degrees - typically,
what a 6-inch focal length camera can record on 5-inch film - will cover only
about 225 feet on the ground when flying at 300 feet altitude. To extend the
cross-track coverage, a "fan" of cameras must be used, or panoramic cameras must
be employed. Panoramic cameras using existing lenses can give side-to-side
coverage from horizon to horizon.
Another factor directly affected by the angular coverage is the cycling
rate of the camera equipment. The tendency has been to make tactical aircraft
fly both faster and lower to make them less vulnerable to enemy defenses, which
then requires, for any given angular coverage, higher frame rates. Frame rates
in modern reconnaissance cameras range from one frame every 1 to 3 seconds for
the larger formats to 3 to 8 frames per second for the smaller formats. These
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67BOO657ROO0300210001-4
SPECIAL HANDLING
Table 2-2 - Angular Coverage of Different Focal-Length Lenses
on Standard Film Formats
Focal
Length
Film Size
(inches)
70-mm
5-inch
9 1/2-inch
1.5
73,7?
2
_58.8
96.8?
3
41.1
73.7?
6
21.2
41.1?
73.7?
12
10.8?
21.2
41.10
24
5.4?
10.8?
21.2?
Table 2-3 - Equivalent Ground Distance Covered (in feet)
by a Vertical Camera of Given Angular Coverage
from Several Representative Altitudes
Angular Coverage
Altitude (ft)
300
1000
2000
30,000
50,000
73.7?
450
1500
3000
45,000
75,000
58.8?
340
1120
2240
33,700
56,000
41.1?
225
750
1500
22,500
37,400
21.2?
110
375
750
11,300
18,700
10.8?
_
375
5,600
9,300
5.40
2,800
4,700
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67BOO657ROO0300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w W
SPECIAL HANDLING
rates are barely adequate for current low-altitude missions, and may fall short
of projected requirements.
Table 2-4 is a compilation of cycle times required in order to obtain 55
percent overlap (for stereo coverage) in cameras of varying angular coverage
operating at representative altitudes, typical of both the low and high-altitude
missions. For each case, three values have been computed, corresponding respec-
tively to 700 feet per second (415 knots), 1000 feet per second (590 knots), and
1500 feet per second (890 knots). Even using a relatively wide-angle lens cov-
ering 73.7 degrees-typically, a 3-inch lens on a 5-inch format - operation must
be at five frames per second when operating from 300 feet at a ground speed of
1000 feet per second. At smaller angular coverages, required cycle times become
shorter; a 41.1 degree coverage from an altitude of 300 feet and at a relatively
moderate velocity of 700 feet per second results in a cycle time of only 0.14
seconds, which is equivalent to a frame rate of 7 exposures per second.
Thus, the frame rate (i.e., the inverse of the cycle time) increases direct-
ly with the ground speed (V), and decreases with altitude (h). It is common
usage to say that the frame rate is proportional to the ratio of V/h. V/h ratios
can be expressed either as feet-per-second/feet-of-altitude, in which case the
resultant dimension is sec-1 or radians-per-second; or it may be expressed as
knots/feet-of-altitude or knots/foot. As an example, the values shown in Table
2-4 range from 700 feet per second/50,000 feet of altitude to 1500 feet per
second/300 feet of altitude, which is equivalent to 0.014 to 5.0 radians/second.
Alternatively, it can be said that the V/h ratios range from 0.0083 to 2.96
knots/foot.
Camera size is influenced primarily by the focal length of the lens and by
the film size employed. A 12-inch focal length camera recording on 9 1/2-inch
film is the equivalent in coverage and frame rates to a 3-inch focal length
camera recording on 70 mm film; furthermore, if the larger camera averages
around 15 lines per millimeter while the smaller one yields 60 lines per milli-
meter, the two will be equivalent to one another in the amount of information
obtained. Yet, the first camera will easily occupy 12 to 20 times as much
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Table 2-4 - Cycle Times Required at 55% Overlap for Various
Angular Coverages, at Representative Altitudes and Ground Speeds
Angular Coverage Altitude
700 ft/sec
Ground Speed
1000 ft/sec
1500 ft/sec
73.7?
300 ft
0.29 sec
0.20 sec
0.14 sec
1000
0.97
0.68
0.45
2000
1.9
1.4
0.90
30,000
29.
20.
14.
50,000
48.
34.
22.
58.8?
300
0.22 sec
0.15 sec
0.10 sec
1000
0.73
0.51
0.34
2000
1.4
1.0
0.68
30,000
22.
15.
10.
50,000
36.
25.
17.
41.1?
300
0.14 sec
0.10 sec
0.07 sec
1000
0.48
0.34
0.23
2000
0.96
0.68
0.46
30,000
14.
10.
6.8
50,000
24.
17.
12.
21.2?
300
0.07 sec
0.05 sec
0.03 sec
1000
0.24
0.17
0.11
2000
0.48
0.34
0.23
30,000
7
5.1
3.4
50,000
12
8.5
5.6
10.8?
300
1000
_
2000
0.24 sec
0.17 sec
0.11 sec
30,000
3.6
2.5
1.7
50,000
6
4.2
2.8
2-8 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
space, and will weigh over 10 times as much as the second one. The obvious
questions are why aren't we using only small cameras, and what are the advan-
tages of the larger cameras over the smaller ones?
The first question can be answered by saying that initially space was
available to accommodate the larger cameras. However, the trend throughout the
tactical reconnaissance field has been away from the 9 1/2-inch format and to-
ward the 5-inch and even 70-mm sizes, the reason being that there just is not
enough room in the modern high-performance tactical aircraft to accommodate the
larger cameras. This is even more true of unmanned reconnaissance drones which
nowadays tend to be smaller and less vulnerable. Here 70-mm equipment has come
into much use.
As for the second question, the main advantage of 9 1/2-inch photography
over the smaller sizes is that it can be viewed directly without the need for
making enlargements or using magnifying viewers. However, where hard-copy
prints must be distributed, modern techniques make it almost as easy to print
enlargements as it is to make contact prints. In the final copy, it is often
very difficult to say whether the print is a 4X enlargement from a 70-mm frame
or a contact print from a 9 1/2-inch film.
As a whole, low-altitude cameras are relatively small while high-altitude
cameras tend to be larger.
One of the most important factors in determining the size of a camera sys-
tem is the total area one must cover in a single mission and, in turn, the quan-
tity of film that must be carried. It can readily be seen that flying low and
fast requires the exposure of a great deal of film in a very few minutes. For-
tunately, the fine detail obtainable from this type of low-altitude mission is
not usually required over more than a relatively small area.
Typically, a 3-inch focal length camera using 5-inch film and flying at
1000 feet will use as much film (approximately 140 feet with 55 percent overlap)
in covering a 40-mile long strip, as will a 12-inch focal length camera also
using 5-inch film, but flying at 30,000 feet and covering a strip 300 Nautical
Miles long. Also, the high-altitude camera will cover this strip with a swath
over seven times wider than the single low-altitude camera.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
t- I _T
SPECIAL HANDLING
So far, the discussion has been based primarily on the use of frame cameras.
They are the most widely used type of camera, and until quite recently were
practically the only ones available. However, as pointed out in the discussion,
modern trends in tactical reconnaissance has promulgated two basic problems:
1. The need for ever faster cycling rates when using frame cameras at
continuously lower altitudes and faster speeds, and
2. The need for wider cross-track coverage than can be obtained with con-
ventional lenses.
These problems have led to the development of the Strip Camera and the
Panoramic Camera.
The strip camera developed from the concept of image motion compensation
(IMC). The basic idea of IMC is that photography of a moving object with the
maximum amount of sharpness requires that the object's motion be followed during
the time of exposure. Press photographers have done this for years when taking
pictures of automobile races and similar events; they track the object to be
photographed, and operate the shutter some time during this tracking operation.
In reconnaissance photography, where the ground appears to move under the
aircraft, often at a rather fast rate, IMC becomes important particularly when
ambient light conditions do not permit operation at fast shutter speeds. Most
high-performance aerial cameras have some means of image motion compensation.
Although a number of ways exist to "track the ground" during film exposure, one
of the better methods is to move the film inside the camera in proportion to the
apparent ground motion. Done properly, this means that although the image of
the ground moves inside the camera, the film on which the scene is recorded will
move at precisely the same rate, resulting in a sharp, unblurred photograph.
The ratio required for perfect IMC between the apparent ground motion and the
film motion is focal length over the flying altitude above terrain.
The next step in IMC development was caused by the following reasoning.
Since film is transported between frames, the transport mechanism could be made
2-10 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w w
SPECIAL HANDLING
to be double duty. If the camera is oriented such that the take-up reel is in
the direction of flight, then a relatively small amount of film can be trans-
ported during exposure (and at the proper speed) for IMC. After completion of
the exposure, the balance of the exposed frame could be taken up.
After this, it only remained to combine the idea of IMC with that of a focal
plane shutter, in which the total format is exposed by a slit traveling from one
edge to the other. The width of the slit divided by the rate at which it moved
across the format determined the exposure time. This became then the strip cam-
era as shown schematically in Fig. 2-1.
Using Fig. 2-1, a lens of focal length f is shown being flown over terrain
at an altitude, h. The ground appears to recede at a velocity Vg, and therefore
the image in the camera moves forward at a velocity Vg x f/h or f x Vg/h, which
is the V/h ratio that we had encountered earlier. To compensate for this image
motion, the film must be moved at this same rate in the forward direction. If
a slit is inserted between the lens and the film and its position is fixed with
respect to the lens, we will have a continuously operating focal plane shutter,
with the difference that now the slit is stationary and that the film moves.
Again, the exposure time will be determined by the width of the slit divided by
the rate of film motion.
The advantage of such a system for high V/h rates, i.e., combinations of
high velocities and low altitudes, is that it voids the need for increased
cycling rates; the entire film supply is exposed in one continuous IMC motion,
and no intermittent shutter or film transport action is required. Itek and
other companies have built such cameras capable of operating at V/h ratios of
six radians per second or more. This is equivalent to operating at Mach 1 from
200 feet. The problem with this camera lies more in the opposite direction. In
order to provide reasonably short exposure times at very low V/h ratios, one
must reduce the slit width to only a few one-thousandths of an inch. Also, the
strip camera does not solve the problem of cross-track coverage, which still is
limited to what a normal lens can cover.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Vg
Fig. 2-1 - Strip camera schematic.
2-12 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
MW
The panoramic camera, like the strip camera, is a scanning type camera,
i.e., it only looks through a slit at a small strip of terrain at one time.
Unlike the strip camera however, which sweeps the terrain in the direction of
flight, the panoramic camera sweeps the terrain from side to side across the
line of flight.
A typical example of a panoramic camera is the Hyac type. As shown in
Fig. 2-2, the design involves rotating the lens about its rear nodal point and
accurately positioning the film in the proper focal plane location by means of
two small rollers mounted on the end of a scan arm which rotates with the lens.
The scan arm also carries the scanning slit. With the lens sweeping at a given
rate, the width of this slit (in the direction of scan) determines the amount
of exposure given each picture element; the wider the slit, the longer the ex-
posure. In a typical application, this width may be varied from 0.07 to 0.40
inch. The length of the slit across the scan direction must cover the width of
the image strip to be obtained. The maximum scan which can be obtained from a
camera of the type shown is necessarily somewhat less than 180 degrees, since
at that angle the lens would be looking back onto the film platen. (The ex-
ample illustrated here was designed for 140 degrees of angular coverage.) How-
ever, as can be seen in Fig. 2-3, a full 360-degree coverage can be obtained
with a direct-scanning camera if the optical system is offset by means of two
mirrors.
Note that in the direction of scan, only the lens and scan arm move, while
the film remains stationary. Furthermore, since the center of the lens rotates
as a unit with the scanning slit, the sharpest possible image is always projec-
ted onto the film, even if irregularities should develop in the scan rate. At
the very worst, there could be some variation in the exposure because of these
irregularities (resulting in corresponding variations in the film density), but
the resolution would remain essentially unaffected.
Fig. 2-4 illustrates the coverage that can be obtained with the three types
of cameras discussed. It can be seen, that although the panoramic camera
solves the problem of cross-track coverage, the cycling rate problem
remains; continuous cover in the direction of flight is obtained by sequential
scanning at properly selected intervals.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
2-14 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
NW NW
SPECIAL HANDLING
Fig. 2-3 - Direct scanning camera with 360? coverage.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Direc tion
of Flight
Direction Direction
of Flight of Flight
Fig. 2-4 - Ground coverage obtained with (A) frame, (B) strip, and
(C) panoramic cameras.
2-16 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Having discussed some of the characteristics of aerial reconnaissance
cameras, Table 2-5 presents a listing of some representative types of aerial
reconnaissance cameras. No attempt will be made to show a complete listing of
these cameras, but rather the intention is to show the more recent models, and
primarily those that are currently in use.
The listing is arranged by camera types and the film width used, and
includes 27 cameras. For a more complete compendium of existing cameras, refer
to the catalog published last year (1964) for Recon Central, Reconnaissance
Applications Branch, Air Force Avionics Laboratory, Wright Patterson Air Force
Base, Ohio, under Contract AF 33(615)-2020. The Contractor was Data Corporation,
7500 Old Xenia Pike, Dayton 32, Ohio.
The problem of IMC which compensates for predictable image motion has been
discussed briefly. Random image motion, primarily caused by aircraft perturba-
tions, may also cause considerable image degradation. In general, these degra-
dations are of less concern for the low altitude case. At low altitudes, reso-
lutions are sought on the order of one foot at altitudes of 1000 feet, or in
terms of angular resolution, about one milliradian. Thus, even with a relative-
ly high roll or pitch rate of 4 degrees per second (70 milliradians per second),
the use of a slow exposure time such as 1/250 of a second is certain to keep the
image blur at less than 0.3 milliradians. Slower exposures will reduce this
still further.
For the high-altitude missions, where typically a 2 to 3 foot-resolution is
sought from 30,000 feet or more, the aim is for an angular resolution of 0.1 to
0.05 milliradians and smaller. Using the same numbers as in the example above,
the image blur (0.3 milliradian) caused by such roll or pitch rates would amount
to several times the desired resolution. Under these circumstances, a stabilized
mount would have to be used.
In its simplest form, a stabilized mount suspends a camera in a gimbal
arrangement which has such low friction that any torque applied to the system
will act on the gimbals, leaving the camera undisturbed in the center. The
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
C
M
N
N
d
C
M
m
N
y
e
O
co
00
C
O
w
0
++
0'
~J
u
03
.
..a
10
{CO
cc
0
0
O M
C6
w 4)
N
0.
N U
C1
p.
L 0
?r.
N N
t4 C 4)
40 U M C
-4 41 W) w
0.. M A. F.
C). 4) N ?:t .t O
> M H t+++
.,4
4) rl L N
40 U N . N
o () C C
'O 40 )+ +. N "4 9) 41
B'0 000 4040W 4)O C C
00 a.0 0. 0. 0 > 40 4) 0
>1 P, P% 0 0.0 0. u 0 CIO
to 46 W 4) k x 4) 0. CO 0) 4..M 4 %t
Q RAo p. O) $' Z'.E. H M N33
d }. U LO Y+ ??? ???
y u j L
0. C.) t- OD c) cq $4 . ? to ? c) 04 C C
0
C+. L- M tD .? tD tD tD tD MO V' O .-4 .-i*-4.-+ U U U
4)
U
' d
4. O O
v I!')O OtD
P. N to to
000
0000
0000 PRW4~
U
C C
W N
C + C O)
"4 w G M
N ??
C l - ..t N 00
?.44 .,
w N
tD SD w tD w
O 10!) 0 0 0 0 0 0
Q) N 0 0) 0) 0) 0) C)
CO - U90)MMMM
M M M C C
?.a. .N SD u? ti to vC?I ?.C. M ?.Ci ?.CI
40 40
M N M ?-).-+M tD-N t0M4000 tD co
N 7N7
O O
O D
C C
Lf) LO
co
RI N 4) N
?0 +) ddd vv-b _'v v d d
?.+ . + pop ~0p 40p ^.t .+ ?.4 a) w ?.+ pop 0
U N Q CO t0 CO C C C U U U 4+ U U N 44
H .Y 4. V U U O 0 0 W H ). 0 H )i u .~ U
?.i d J 'C M wd v) U U U M M +.?1 N M ?.i r) 67 ?.t
.0 L .. .C .C .C >, >, >+ ib Q! v A7 RI i.) T
is ~?E0 UUUS SS wG..wk.w 1-40
N tD OOV.-
dd1 1
.6
0 dddN
S+G AL AL bL ~ ~
aI ) 1 I ) )
~G ~ ?U4 bUG pUG ?U4
SPECIAL HANDLING
N M
N N
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
co
co v
w cv L() cq
Drachm to r-1
sa
w
a o
ro,-. o
v o
-4
w o 0 0
?~~. ooo0Ln
LO LO ca
0 0 0 0
O O O O
d W W 1-4
H H r-I r?-I
SPECIAL HANDLING
0 ooo
0 0 0 0 0 0
0
-4 LO 0 NLO 0 LO 0
-4 Cl
0 0 0 0
O 0 O O O
00 O 00 00 00
_ M H , -I '-4
?~ C C
G 14 C
?, rj
C4
00 Gq
r-4 M CO 00
r-4 M
-4
N
v ro
N
v
r-1 H
FT7 -4
N ?rl la ?'a H U .,4 }.1 .14
Hwpaw W W paw
H LC -i Q Q
U Lf) LC) to -4 L() LO LO M
w0~ ~0
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
higher the moment of inertia of the camera and the lower the gimbal friction,
the more stable will be the camera in its mount. Ideally, with zero friction,
no torque can be coupled through to the camera, and the camera would be com-
pletely motionless regardless of any rotational disturbances to the gimbals.
However, in actual practice idealized conditions can only be approached.
Since the primary function of a stabilized mount is to hold the camera steady
during exposure, the criterion for the quality of this type of mount is the
degree to which it can hold a position while an aircraft is going through typ-
ical roll, pitch and yaw motions. Typically, these aircraft perturbations may
occur at rates of several degrees per second. A good mount can reduce this
motion to 7 seconds of arc per 1/100 second of time, and mounts that are steady
to 20 seconds of arc per second of time are possible.
In terms of ground resolution, the image blur resulting from a camera
motion of 7 seconds of arc per 1/100 second of time when photographing at an
exposure of 1/300 second from an altitude of 45,000 feet corresponds to only
about 6 inches. The image blur of an unstabilized camera under similar flight
conditions and assuming a 4 degree per second aircraft motion would correspond
to over 10 feet. The main disadvantage of a stabilized mount is its size and
weight. Such a mount will weigh from about half to over twice as much as the
camera it carries.
Although it has been recognized that modern warfare, and especially
guerilla type warfare, will rely increasingly on night operations, the develop-
ment of night photographic systems has lagged. The primary reason is the diffi-
culty in providing illumination, even over a restricted area, that compares with
the light that the sun provides during daylight hours. The efforts have there-
fore included not only the search for bigger and more powerful light sources,
but also the search for ever more sensitive materials and sensor systems which
can take advantage of what moon or starlight is available.
In the area of artificial illumination, the primary emphasis has been to
provide brief, high-intensity flashes corresponding to the flash bulbs used by
press photographers. Two such types of artificial illuminants are:
2-20 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
1. The pyrotechnic cartridge and,
2. The electronic flash
2.7.1 Pyrotechnic Cartridge
The pyrotechnic cartridge is usually ejected from the aircraft carrying the
camera equipment, and is fused to ignite from 0.6 seconds to several seconds
after ejection. Depending on their size, these cartridges furnish enough light
to properly expose a fast film in existing cameras at altitudes from 1200 feet
to around 4000 feet. (A number of flash-bombs exist for single shot operation;
these weigh as much as 165 lbs and are capable of illuminating the terrain below
from up to 45,000 foot altitude.) This is one of its advantages over the elec-
tronic flash systems which generally can furnish sufficient illumination only
for altitudes of 1200 feet or less. Other advantages that pyrotechnic cart-
ridges have over electronic flash systems are smaller size, lighter weight, and
negligible power consumption.
For practical purposes, three types of cartridges are presently in use:
the M-112, the M-123 and the XM-161 (Daisy) cartridge. Their characteristics
are compared in Table 2-6. The table shows that for low-altitude night mis-
sions, several dozen of the small cartridges can be carried with an added weight
comparable to that of the camera itself. The M-112 cartridges, weighing 1 pound
each, will not add excessively to the weight or space requirements of the sensor
package. Power required is only that necessary to ignite the ejection charge.
However, pyrotechnic cartridges have serious drawbacks which are as follows:
1. The flash duration is approximately 15 milliseconds for. the smallest
and as long as 40 milliseconds for the largest of the cartridges. This corre-
sponds to exposure times between 1/65 to 1/25 second which are very long in
terms of use for aerial photography. So unless the camera is stabilized, and
equipped with excellent IMC provisions, considerable image blurring may result.
2. Most aircraft have a limit on the number of cartridges they can carry
which limits the number of photographs they can take at night.
3. Cartridges are expendable ammunition that must be replenished after
every mission. The explosive character of these devices make them very unde-
sirable to operate in training, in peacetime, or over friendly areas.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
C 0
05
) N L U
C
U
.~~ N U)
(0 -
N
N
C
C
0.
.-a
0
0
-
?.M
W
?+
t
In
CO
0
CO
N
0
.3
.-I
-
.-I
CO
A
?=I
so
I
N
0
>4
.--I
-
Cr
aCi
W
?fl
N
It7
O
C
- 4
C?
h
d'
CO
d'
.-+
CO
N
R
C" )
N
.-I
N
.-I
U
+a
u
i+
+~
N
N
C
C
O.
Q
o
0
?.+
v+
w
M
N
M
L"
IE
s
U
.t
't
4
e}I
(
O
N
[ .
N
If)
.+
O
.-I
u
e0
4J
L
'4
u
++
w
0
O
a,
N
s
-
d
v
O
1+
>
Z
N
a+
L
ar
N
.C
60
t
0
u
C
a
1
N
M
3a w 4) a
d
- . e
w a
_
P-4
>
N
.O
N
CO
CO
C
.C
.O
O
-I
. .4
Co
II
4.
O
aa~
>
O O
U
C)
a~
m
'C
10
.0
V'
.A-i
dI
O
N
CO
N
COD -4
V
N
U)
.O .O
~4 -4
N
N
N If)
r+ .4
40
.-i
d
U)
U') O
2-22 SPECIAL HANDLING
_ -4
10
N C C
C N 10
I 8 b
-~ a C
.o d d a 10
4
Q u 0 0
0 I) O O
O O O co
) O O NCO')
C
10
C u
0
++ u u
.-i N (4 O
8
~O~II O
It) O O Co
N O O to
C
10
-C U
0
- 0) 61 O
H O
O C9 O
O O O dI
.d O CO
..I
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
2.7.2 Electronic Flash
Electronic flash equipment has the problems of being cumbersome, heavy and
requiring considerable power (up to several kilowatts). The main problem arises
from the need for storing up to several thousand joules (or watt-seconds) of
electrical energy in banks of capacitors so that it can be dissipated, almost
instantaneously, in the flash tube. Power supplies, capacitors, reflectors, and
associated cabling weigh about 200 pounds. The result, however, is a very short
burst of light - typically of the order of 1/10,000 of a second. This short ex-
posure time is the primary advantage of the electronic flash light over any other
night-time sensor system. This very short exposure effectively stops all air-
craft or ground motion. Such a system needs neither IMC nor camera stabilization.
The principal drawback of the electronic system is its limited light out-
put. A demonstration was given in the spring of 1964 of one of the most power-
ful flash units developed to date. It was flown at a 1200-foot altitude, and
covered an area less than 1000 feet square. Examination of the resultant
photography showed that only about 25 to 30 percent of that area was adequately
illuminated (2,000 joules, 6-inch f/2.8 lens, Royal-X-Pan film).
A number of developments are underway which, by means of "light intensi-
fiers", generate an image on a phosphorescent screen from the very low light
levels existing under moonlight or even starlight conditions. By cascading two,
or even three of these stages, an image of sufficient brightness eventually re-
sults which can be recorded photographically. At this time, none of these pro-
grams have resulted in operational hardware, but they should be watched to see
if they could offer any solutions to the problems of nighttime photography.
The illuminance parameters for typical electronic flash systems are illus-
trated in Fig. 2-5, which indicates the terrain clearance possible with various
film-lens combinations. It should be noted that an electronic system requires a
finite recharge time, thus limiting the power versus the cycle time parameter.
This is reflected in a reduced flash intensity and terrain clearance. The
typical parameters of this relationship are shown in Fig. 2-6. It can be seen
that low light levels, which require low altitude operation, result from high
frame or cycle rates, thus reducing focal length and ground resolution in order
to have sufficient light for exposure.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
2.8 CAMERA CONTROL SYSTEMS
This section on Photographic Sensors is concluded with a brief discussion
of camera control systems. It was shown throughout the description of the
various camera types and auxilliary components that a complete camera system
performs many functions. A few of these are as follows:
Camera system placed on standby
Camera system turned on
Sensing of light level
Calculation of proper exposure and diaphragm settings
Sensing of height
Measurement of velocity
Derivation of the combined V/h ratio
Determination of the frame rate
Setting of IMC rate
Distribution of signal pulses
Monitoring of proper operation
Display of warning lights in case of failures
Computation of data to be recorded
Distribution of information to cameras
In the case of night operation, additional functions are added:
Ejection of pyrotechnic cartridge
Triggering of cameras after a suitable delay
Sensing the flash
Camera shutters are closed, recycling initiated
Selection of next cartridge
Some mechanism must be provided to perform all of these functions in proper
order and, where more than one photosensor is involved, provisions must be made
for the distribution power, control signals and recording data to other cameras
or other reconnaissance sensors. This mechanism which performs these functions
is the camera control system.
Two basic systems are now in general use. The Universal Camera Control
System (UCCS) and the Simplified Camera Control System (SCCS). These systems
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
consist of a number of mutually compatible modules that can be assembled to fit
a large number of different camera systems configurations.
Some of these modules or building blocks are input devices. They permit
the pilot or operator to set desired frame rates and basic exposure parameters,
such as the filters desired for the mission and percentage of overlap between
successive photographs. For night operation, such input modules permit preset-
ting the number of flash cartridges to be released during one photographic
sequence, and the rate at which they are to be ejected.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
t= t
SPECIAL HANDLING
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
VW
SPECIAL HANDLING
\\\ \ \
\ \\
`\\ \\
z auP
4
\\
\ ` \
\
V
`\\ \
0 2
\
\ \
\
~`\\
\\
\
W o w
32
\
\\ \ \
oe
\\\ \\\
ro
~, ~C'
\ \\ \ \
u1
~t
o \ \\ \
~
\
1 \ \\
\ \\
\
op
4
K \~
\ `
~C
\
w \\
\
\ \
I-
NZ` \
\\ \\\
~~~ p ki
`
\\
\
. Q ~jNN1P
c V U
\
\
\\
.
\
\
a
\ ~\
\
claa=,) aamyxva,~) P11b'3213L
SPECIAL HANDLING
W
W
F?-
x
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
3000 6000
ZSOO soon
2000 4000
I TOO 3006
SPECIAL HANDLING
74- KITTS
LGNS
Folic. c.EN4TH
-4-xaz
Ftw Foitm r
No ? SBA LIveL of og o1 (.Z 'a 4 -!o
400
i i 400 i b i xa F t o 0
5Z`/o
,4ttC AP1 69ouMD 'VPeeD- KNOTS -+= --~
SPEC IAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
VOW
SPECIAL HANDLING
Other modules are actuators for 'camera position doors, power amplifiers for
servo motors for camera and film magazine drives, pulse shapers for proper signal
distribution, and junction boxes to tie several camera stations together.
Still other modules are automatic intervalometers, which compute the proper
interval between camera exposures to provide ground coverage at a predetermined
overlap for stereo coverage. Others furnish the voltage required to transport
film at the proper IMC rate during exposure. All of these modules are mounted
on one of several standard base plates, which are equipped with mounting pro-
visions, power and signal connectors, vibration isolators, etc.
Some additional control components have been under development, for some
time, such as direct image motion sensors to provide accurate IMC signals by
direct measurement of the apparent ground motion rather than to derive these
from navigation data. Other devices provide accurate vertical indication which
is important in mapping and charting. Also, improvement is being made on data
recording instrumentation to provide aircraft altitude, and other flight data
as well as time and even geographic longitude and latitude information for direct
recording on each photograph.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
3. PHOTOGRAPHIC DATA PROCESSING
Upon completion of the photographic mission, its value then lies in the
delivery of the photographic material to the user. However, before the user can
extract any information the material contains, it must be processed-with regard
to time so as to maximize the validity of the information, and with regard to
quality so as to maximize the content of the information.
Processing can be accomplished during flight or at a ground processing sta-
tion. Inflight processing approaches real-time availability; however, it does
not present the best quality. Ground processing provides high quality, but re-
quires additional time over inflight processing. Rapid processing techniques
are being improved to narrow this time margin. In addition, duplication proc-
esses can be used to enhance the information content.
This section deals with all of the above facets to familiarize the reader
with the technology available in photographic data processing.
3.1 INFLIGHT PROCESSING
A number of techniques have been evolved in recent years to provide in-
flight photographic processing including those utilizing free liquids and ab-
sorbed liquids.
These devices and techniques have, for the most part, satisfied the partic-
ular requirements of the system for which they were designed. How well each
technique might satisfy the general parameters of airborne rapid processing will
now be discussed, including the major advantages and disadvantages inherent to
each approach.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
3.1.1 Atomized Spray
Very rapid processing has been accomplished by subjecting the photographic
emulsion to a heated spray of developer, fixer and rinse bath in sequence. Var-
iations of this technique include single and multiple spray nozzles, and con-
tinuous or step and repeat processing.
In the Kelvin-Hughes RP series, a Venturi-type jet operated by compressed
air is used to bring the atomized chemicals in contact with the emulsion, with
processing being accomplished at rates up to 40 feet per minute in some models.
Solutions are used only once, and then discarded. Processing temperatures are
usually in the order of 100?F, providing for very rapid development and short
access time. Densities of 2.0 and gammas of 1.3 have been achieved by this
process with Ilford BY film in approximately 10 seconds total processing time.
While this system can provide rapid processing and relatively consistent
quality, there are serious problems inherent in the handling of the liquids and
the subsequent spray condensate. The situation is further compounded by the
fact that the size of the spray droplets effect the uniformity of the processed
image.
Such a system, being potentially subject to leakage about the film seals,
solution container, and other points, greatly reduces its application for air-
borne usage.
3.1.2 Viscous Processing
Several techniques of utilizing viscous processing solutions have been in-
corporated in airborne processors in recent years. Common to such systems is
some type of mechanical dispenser which applies the viscous chemistry directly
to the film emulsion. The chemistry emitter may be a monobath for single stage
operation, or a dual system applying developer and fixer sequentially.
In the Fairchild PTS system, a plastic overlay is provided encapsulating
the processing chemistry and permitting immediate film handling and evaluation.
Other models, like the Fairchild KS-64, provide for the removal of the viscous
processing chemistry along with the plastic cover, leaving a damp emulsion
3-2 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
suitable for readout or high energy drying. The use of the highly thickened
solutions (up to 60,000 centipoise) greatly reduces the possibilities of equip-
ment damage due to leakage of the chemistry, but restricts the mobility of solu-
tion and retards chemical activity. Although access times can be obtained as
low as 10 seconds with certain combinations of film and monobath, approximately
1 minute total processing time is required for two-stage systems with films such
as EK type Tri-X aerial at raised temperatures.
Such viscous processing systems are advantageous for airborne processing
principly because they are relatively attitude and altitude insensitive, and
access times are minimal. Solutions being used only once assures uniform qual-
ity, and the sensitometric characteristics approach those of conventional proc-
essing, although with some loss of speed.
The use of a mechanical applicator, in which close tolerances are required,
and elevated temperatures necessary for rapid readout capability, result in the
possibility of film abrasions and scratches as well as jamming or breakage of
the film at the applicator station. Maintaining the proper orientation of the
plastic overlay on the film during subsequent readout or handling requires spec-
ial care, and drying upon removal of surface chemistry by preferential separa-
tion is very difficult due to residual chemistry in the emulsion.
The use of various types of chemically saturated webs have been employed in
airborne processing systems with varying degrees of success. Plastics, paper,
foil laminates and film base have been utilized as the carrier.
These systems offer simplicity of operation, compactness, and attitude in-
sensitivity. The mechanics of the operation require only the contact of the
previously saturated web material with the exposed film. This is usually accom-
plished by mating the two materials through a pressure roller system, the web
being drawn off its supply spool as required.
The Eastman Kodak Bimat system produces a positive image of the negative in
process on the carrier web which, in this case, is a gelatin coated polyester,
using monobath chemistry and diffusion transfer. Both the negative and positive
are usable after processing although neither film is dry.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Limitations of the system are the relatively short life of the imbibed
solution, and the detrimental effects caused by premature drying of the solution,
which results in voids or unprocessed spots where contact was imperfect or solu-
tion was not fully absorbed.
Some speed loss is incurred and density range is lowered, but maximum re-
solving power is retained. The requirement of a 15 to 20 minute, 80?F process-
ing time of the negative does not rank this system with the fastest, but the
positive image can be obtained in approximately one minute.
The Itek web system utilizes a porous plastic web saturated with a high
energy monobath. This web produces a high quality negative and is capable of
being modified to produce a positive print on the web.
Utilizing a very versatile monobath, the web processes EK type 4404 film in
less than 5 minutes, Ilford Aerial-N in 10 minutes, and EK type Royal-X Pan re-
cording in 15 minutes, all at 70?F. Maximum densities of greater than 3.0 have
been obtained on some films.
Saturated web processing techniques have in common the advantages of sim-
plified solution containment, relatively failproof mechanics, and independence
from attitude. The disadvantages are primarily storage life of imbibed solu-
tion, evaporation of chemistry during operation, and limited mobility of
solution.
3.1.4 Liquid Cell Processors
Optomechanisms Inc. has developed an airborne processing device which util-
izes a "liquid cell" directly in contact with the emulsion of the exposed film.
The entire exposed frame is in a seal as the appropriate processing solution is
pumped into the chamber and drawn off by vacuum. Developer, fix and rinse solu-
tions are programmed into the chamber independently. The film is required to
maintain its position during a three second bath period in the cycle, and is
therefore not suitable to continuously moving film. The problems of liquid
handling and transport are further compounded by the reliance on the film as a
perfect liquid seal during the entire cycle.
Another type of liquid cell device is found in a CRT recording processor
developed by the Army Signal Corps. The film is indexed into a separate
3-4 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
1W
*KOO
SPECIAL HANDLING
developer chamber where it is simultaneously exposed through a "window" and lens
system, and developed. The cell is constantly maintained full of developer sol-
ution during operation by a positive pressure pump. Indexed to a similar fixing
cell immediately after developing, the image is projected for interpretation
while clearing. The total time, using Ansco Telerecord film, is less than one
second at approximately 200?F. Leakage of solution during processing is con-
trolled by negative pressure pumps.
The very short processing cycle is partially due to the high temperature
processing, thus limiting the selection of emulsions to those capable of with-
standing such temperatures. Further, the orientation of components (lens, film,
chamber) requires that the film be exposed through the base. This, besides add-
ing distortion, and theoretically reducing speed, would further limit the selec-
tion of films to those having a suitable transparent base.
As with other free liquid systems, the problems of solution leakage limit
this technique to very specific applications.
3.1.5 Viscous Diffusion Transfer
The viscous diffusion transfer system most notably utilized is the Polaroid
Land technique. This offers a relatively "dry" process which is a distinct asset
to airborne photographic processing. Basically, the system consists of the ap-
plication of a developer that contains a silver halide solvent to an exposed
photographic emulsion, while in contact with a second film or paper support that
is not light-sensitive. While a negative image is being formed in the light
sensitive film, some unexposed silver halides are transferred to the nonsensitive
support forming the positive image. This process can be completed in as little
as 10 seconds at 70?F with specially prepared emulsions.
While the technique is not limited to the use of viscous chemicals, proc-
essors have been fabricated specifically for airborne operation using both
rupturable pods of highly viscous processing chemicals for intermittent process-
ing, and piston actuated supply containers for continuous processing.
The system is limited to the use of special emulsions and therefore cannot
be directly compared with other rapid processing systems compatible with a num-
ber of standard emulsions.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Quality, reliability, and environmental requirements are similar to those
of other viscous single stage processes discussed under "Viscous Processing".
3.1.6 Meniscus Processing
A rapid film processor utilizing a porous teflon felt roller with a sin-
tered metal core has been developed by Eastman Kodak to record airborne side
looking radar line scan displays. The device utilizes a liquid monobath at
temperatures in excess of 100?F, supplied through the periphery of a driven por-
ous roller in close proximity to the exposed film. A meniscus is formed at the
tangent point allowing transfer of chemicals to the film without physical con-
tact of the roller and film. Complete processing has been accomplished in 18
seconds at 130?F using EK type S0-1188. The porous roller, being driven inde-
pendently, offers good agitation to the process and the roller-film separation
eliminates the possibility of scratching or binding at the applicator.
The primary objection to airborne usage is the containment of the solution
during the processing cycle. Maintenance of the proper meniscus would be diffi-
cult with a change in the vehicle attitude and with the shock and vibration
characteristics inherent in an airborne environment.
A variation of the meniscus type of processing is found in a unit designed
by Photomechanisms, Inc. The exposed film is positioned against a backup plate,
and a meniscus of the low-viscosity processing solution is formed between the
film and a sintered or porous plate. The processing solution is fed through the
porous plate from a heated solution chamber. Excess solution is collected in a
surrounding trough and routed to a solution waste container. At a processing
rate of 10 inches per minute, access time is in the order of 20 seconds.
The system, utilizing low-viscosity solutions within an overflow circuit,
encounters the difficulties found in other free liquid systems, and is thus not
particularly suited to an airborne application.
3.1.7 Systems Under Development
Research is currently underway by various agencies to simplify and improve
the rapid airborne production of photographic records.
3-6 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
These include the use of dry chemicals in the production of aqueous solu-
tions only when required at the processing station. This would impart almost
indefinite life to the stored processing materials and minimize solution hand-
ling problems. Another approach is a multicell canister of dry ingredients and
liquid, which are automatically mixed and applied to the film as needed. This
system also offers a maximum shelf life.
Other systems involving specially prepared emulsions, including a number of
non-silver halide materials are being studied in view of airborne processing
applications. Detailed information regarding the success of these techniques is
not currently available.
Color - Airborne processing of color materials usually requires a much
greater complexity of mechanics and a subsequent increase in size and weight as
compared to black and white processing. For this reason, only relatively little
work has been done in this field.
The recent introduction of "Polacolor" offers the potential of color by
diffusion transfer for those systems that can utilize a reflection print for
information analysis. This method is likely to see expansion into color trans-
parency production. The system would be subject to the same limitations of
black and white diffusion transfer processing.
A color system utilizing low-viscosity liquids was developed by Kelvin-
Hughes in a modified RP3 unit specifically for the processing of a two color
Ilford radar color recording film. The cycle consists of developing, bleaching,
washing and drying applied through a plurality of jets. Twenty second process-
ing has been accomplished at 104?F. This, however, is a "false" color system
utilizing a specially formulated emulsion made specifically for recording CRT
traces.
Almost all of the previously described black and white processing systems
could be adapted to color processing by duplicating the basic processing station
to handle all the solutions required for color processing. This would result in
a substantially larger and heavier device, and would multiply the areas of poten-
tial malfunction as well as greatly increasing required access time.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
F
SPECIAL HANDLING
Positive versus Negative - Photographic information is generally available
more rapidly in a negative form and photointerpretation personnel have been able
to successfully utilize this type of image. Reversal to a positive image is
possible but always involves an additional operation although this may be accom-
plished simultaneously with the negative production (Bimat, diffusion transfer).
All of the processes previously mentioned produce a negative image unless
otherwise noted.
Base Coatings - Most emulsions currently available are manufactured with
some type of film base coating. In some cases, as with EK type 4404 film, a
colored dye backing is normally removed during wet processing. However, rapid
processing systems are generally arranged such that only the emulsion side of
the film is treated in order to conserve chemicals and to facilitate drying.
These systems are therefore limited to use with those emulsions whose base coat-
ings do not restrict printing or visual observation.
3.1.9 Summary
The ideal airborne photoprocessing system would have the following features:
1. absolute reliability
2. real-time readout
3. minimal weight and size
4. no safety hazards
5. freedom from attitude and altitude
6. low power requirements
7. dry to dry film handling
8. acceptance of wide variety of emulsions
9. acceptance of color and black and white film
10. maintenance of high sensitometric, optical and physical characteristics
11. simplicity of operation and maintenance
12. indefinite storage life of processing agents
13. archival life of processed film
All these factors are important only as they affect the successful comple-
tion of the objectives of the particular program. In many cases only a few of
these items are of importance to a particular system, and one factor may far out-
weigh all the others. The type of airborne processing selected for a system,
therefore, depends upon the specific requirements and parameters of the program.
3-8 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
%W %W
SPECIAL HANDLING
Where access time requirements are of more importance than quality, high
temperature spray monobath may be the optimum technique, but certain safety
hazards are involved here as previously noted.
Optimum safety considerations would make a high-viscosity process more
acceptable than some other methods. Demands of simplicity, light weight and
relative freedom from malfunctions would merit the study of a web processing
system, particularly if a colored dye backing is involved. For the immediate
production of positive imagery, diffusion transfer or a double web system would
be most satisfactory.
There is currently no one system that offers a complete solution to all the
problems and answers all the requirements of airborne photoreconnaissance.
Although inflight processing yields close to real-time data and informa-
tion, ground processing remains the best way of handling reconnaissance photog-
raphy to produce maximum image quality and optimum tone reproduction. The in-
telligence yielded by high acuity reconnaissance photography depends greatly
upon the quality of the processing procedures. Today's photographic reconnais-
sance systems are able to record vast amounts of information. Modern films and
lens systems have been optimized to a point where they can be deleteriously
affected by the ground processing and printing operations.
To be able to preserve this information for later critical analysis, the
processing of the original reconnaissance film must be carefully done. Local-
ized processing nonuniformities, machinery breakdown, abrasions, dirt, uneven
drying and water marks can cause a serious reduction, or even complete loss of
reconnaissance information.
Since a photographic record consists of an infinite number of density
changes, any alteration of the true density record through non-uniform process-
ing may cause either loss or alteration of the recorded information. Much
reconnaissance is done under adverse weather conditions and at extreme altitudes.
Constant controllable sensitometry from day to day is essential to accommodate
for these variables.
SPECIAL HANDLING 3-9
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Information is wanted not only of high quality, but also as rapidly as pos-
sible, especially when concerned with tactical operations. In many cases,
photographic processing must be handled in the field far from ready supplies and
under environmental conditions adverse to the photographic process.
Basically, the time to develop a film is constant. The volume of film to
be processed in a given time is the determining factor as to the time required
and the type of processing installation required. For example, a 100-foot roll
of film may be handled easily by a simple reel processing kit, and yield the
same quality as a high cost continuous processor which can process film at 40 feet
per minute, and turn out 20,000 feet of film-in 8.5 hours.
3.2.1 Basic Techniques
Ground processing equipment varies widely in complexity and purpose. Equip-
ment range from the wind-rewind reel type to complex continuous processors which
have automatic development control using IR scanning, spray, automatic tension
control and automatic replenishment. Although a complete list of processors
will not be given in this report, the general classes will be discussed.
Portable Reel Units - Portable reel units (such as the Morse Type 5A) gen-
erally are designed to handle 70 mm to 9 1/2-inch film in lengths up to 250 feet.
These units are self-contained and portable, with agitation provided by the wind-
ing and rewinding action of the film in a static developer. These units have
the obvious advantages of ease of use and portability. Film can be processed in
the field by handcranking if necessary. For precision however, they are hardly
satisfactory as agitation, and hence uniformity and repeatability is poor. The
low film capacity is a disadvantage if larger quantities of film must be proc-
essed rapidly. Generally speaking, the portable reel processors are only useful
for extreme field locations.
Immersion Processors - In common use throughout the military establishment
are continuous type immersion processors which handle up to 9 1/2-inch roll film
on paper. A continuous processing capability is accomplished by splicing one
roll of film onto the end of another without interrupting the movement of the
film through the solutions. There are two basic types of immersion processors -
static tank and immersion turbulation.
3-10 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
IftoF %W
SPECIAL HANDLING
In the static tank processors, agitation is provided by the forward direc-
tion of the film through the solution, and recirculation. Processors of this
type (such as the EH-6) are semiportable, relatively simple to operate and rea-
sonably compact. On the other hand, they give poor agitation which means that
uniformity and repeatability and sensitometric control are relatively poor. Be-
cause of the lack of agitation, the attainable transport speeds are low, having
a maximum speed of approximately 25 feet per minute.
With the immersion turbulation processors, agitation is provided by sub-
merged jets which agitate the solution at the emulsion surface by directing
streams of solution at the film. Such processors (as the HTA-3CM) allow moder-
ate transport speeds up to 40 feet per minute, and good sensitometric control.
Uniformity is good if the turbulation system is carefully designed. For prac-
tical purposes however, a slight streaking often results with this type of
processor. This type of processor can be a good machine with moderate uniform-
ity.
A common advantage with both types of immersion processors is the relative
simplicity of temperature control. The processing system is self-contained, and
the temperature sensing and control can be done in the tank. This type of ma-
chine is somewhat limited from the standpoint of flexibility. Since large quan-
tities of solution are usually required (40 to 90 gallons), it is difficult and
time consuming to change chemistry to do other films. This is a serious disad-
vantage if the laboratory is expected to process many different emulsions re-
quiring different developers.
Spray Processors - For high speed, high acuity processing, spray processing
has proven successful. High pressure spray nozzles are placed adjacent to the
film's path of travel, and the impinging spray removes the reacted solution from
the films surface, replacing it with fresh chemicals.
Film developed under conditions of poor agitation results in the accumula-
tion of a layer of reacted chemicals on the emulsion surface, which must be re-
moved to effect optimum processing. It can be seen in Fig. 3-1 that the inten-
sity of agitation can be increased to a point where no additional photographic
difference can be detected. After this point, the surface layer is so effec-
tively displaced that the chemical reaction is insensitive to moderate changes
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Agitation
Fig. 3-1 - Change in density for five constant exposure values
with changes in agitation.
3-12 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
in agitation. (See Turner, J. R. and Jensen, E. W.; Some Principles of Spray
Processing, SMPTE, Vol. 65, February 1956, page 92.)
The spray nozzles, which generally are in open tanks, can reduce processing
time by a factor of two or three over immersion techniques. At the same time,
lower fog and higher maximum density can result. Processing rates in excess of
100 feet per minute can be obtained with some of these machines. Generally
fresh chemicals are used, thus requiring a continuous chemical mixing capabil-
ity. The logistic support required for one of these processors restricts its
use to large permanent facilities.
The advantages of a spray processor are that good uniformity and sensi-
tometry can be obtained, although extreme care in spray distance and alignment
is necessary. It also produces a high speed and development rate, and the
chemicals can be easily changed to readily accommodate different films. Some
of the disadvantages are high chemical consumption and complex machines which
are prone to failure. Uniformity has been a problem with the wide 9 1/2-inch
films, but the spray jets have worked very successfully for 70 mm film widths.
Self-Threading Roller Transport Systems - Considerable interest has been
given to the new type self-threading roller transport system such as the Versa-
mat. These units are generally compact and portable, and provide a ready ma-
chine for small job processing. They are versatile in that they handle sheet
film as well as aerial film with no modifications. The machines are limited in
the volume of film that can be handled. Agitation is only fair, and emulsion
damage from the 212 rollers required to transport the film is possible. They
also require special chemistry which cannot be used with all films.
Semitrailer Laboratories - Advanced field operations may require the use of
a mobile photolaboratory. Self-contained units have been built being a complete
processing facility built into a trailer. These trailers can either be towed to
the desired location or loaded into a C-119 aircraft and moved by air (see Fig.
3-2). Since these trailers are used in fairly remote areas, problems sometimes
arise in providing the necessary support functions such as water and power. To
compensate for lack of these utilities, ion exchange units and electric gener-
ators can be used.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Continuous
Printer
Fig. 3-2 - Photographic Laboratory ES-14 or FM-25.
3-14 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
NW
SPECIAL HANDLING
Clean Rooms - Clean rooms provide the ultimate in precision processing.
Contamination and dust are controlled to the submicron region. The entire lab-
oratory is sealed off and filtered air is pumped into the facility. Personnel
must observe the utmost precautions in cleanliness and suited to prevent any
body contaminations such as dandruff and skin flaking. By using carefully de-
signed processors and high acuity printing equipment, optimum information can
be preserved in the photographic image. These facilities require the support
of a permanent base operation.
3.2.2 Precision Processing
The trend in recent years has been for extremely high photographic quality
despite the considerable costs involved. The cost of a sortie, the risks in-
volved, and the intelligence information gained considerably offset the cost of
high precision processing. Unfortunately, the setting up of high precision
laboratories in the field for support of local photointerpretation units is not
practical, although considerable work has been done in providing the mobile vans
which are fairly complete with processing and photointerpretation capabilities.
Some field operations are provided with high quality continuous processing
units and the necessary-accessory equipment that must go with it such as ion
exchange, humidity and temperature regulating equipment. Limitations are still
encountered as to the volume of material that can be handled and the general
physical restrictions of limited manpower and supplies.
To the other extreme, clean room processing provides the ultimate in film
processing conserving the film's informational imagery down to the micron region.
Clean room environments require the support of permanent base facilities thus
putting a delay in the acquisition time of the photographs.
Possibly of the greatest importance is the ability to handle great volumes
of high quality photographic reconnaissance. A sortie may produce photographs
of which copies must be disseminated to many photointerpretation elements and
tactical units. The aspect of time is important. Preliminary photoanalysis
may induce return action within minutes. Continued analysis may be radioed to
dispatched units who are traveling with the prints. Complete versatility is
needed With a high volume of quality photographs.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Logistic support of a photographic facility requires considerable planning.
Large amounts of chemicals and utilities are needed. Provisions must be made
for emergency loss of electricity or water. Temperature and humidity of the
facilities must be regulated. Control must be held over the storage and ship-
ping environment of the photographic supplies.
Mission accomplishments bear greatly on the physical design of a processing
facility. High precision processing equipment costs in the tens of thousands of
dollars. Rapid high pressure spray equipment can process film at up to 100 feet
per minute. These machines require a continuous supply of fresh chemicals and
water and maintenance.
3.3 RAPID PROCESSING
The numerous approaches to rapid processing are the natural outgrowth of
the ever increasing need to obtain a visual record of an event with a minimum
of delay. To meet this need, emulsions with improved physical characteristics
and great facility for processing, chemical processes and techniques of increased
efficiency, and specialized processing equipment have been developed.
Airborne rapid processing, a specialization of conventional rapid process-
ing, has evolved as essentially a military reconnaissance requirement. It is
the result of a relatively abrupt combination of electronics, photographic data
recording, camera optics, emulsion technology and process chemistry. The result
has often been achieved at some sacrifice in the potential quality of the record,
but in the majority of cases, the end has been achieved.
In actual use, the term "rapid processing" can be misleading; it may denote
total processing times of as much as 10 minutes, and may even refer to the rate
of travel of a film through a processor, rather than the treatment times in-
volved. A more precise term, particularly for airborne use, is "rapid access
processing", i.e., the length of time required to expose, process and view the
material. This survey considers a process "rapid" if the access time does not
exceed 30 seconds, and ultrarapid if it does not exceed 10 seconds.
This survey presents a summary of airborne rapid processing equipment known
to be available in late 1961, and readily acknowledges that tremendous improve-
ments have been made in this field since that time. Reference is made to an
3-16 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
article published by Hersch and Smith in Photographic Science and Engineering,
Vol. 5, No. 1, January-February 1961, and titled "Rapid Processing: Present
State of the Art".
In reading over the information presented, the following observations
should be considered:
1. The most successful rapid processing systems are the results of care-
fully matching the film, the chemistry, the process, the application technique
and the equipment. The specifications required limit the films suitable to a
relatively few types. The chemistry tends toward high energy, high pH, and high
concentrations of chemicals.
2. While the majority of the processors shown have exhibited airborne
processing capability, only a few have been proven practical and are accepted
by the military in service tests.
3. The units that are commercially available are relatively high in cost,
reflecting low production volume and custom construction.
In the need for immediate access, some systems sacrifice the original nega-
tive for a positive. Others require a post-process to retain the original nega-
tive and to achieve permanency of the positive.
Inflight processing systems are limited usually in size, as to the type
film that can be used, and many suffer losses in emulsion speed, loss in gamma
(contrast), and present a higher fog.
Very few rapid access systems are capable of archival quality, although
granted it is not always required. Most inflight processors add to already
loaded vehicles. The extra load of cameras, electrical systems, etc., must be
carefully considered. Also, it is necessary to consider whether simple viewing
of a positive is adequate, plus scan and transmission via CRT.
Classification of Equipment
A. Directly Applied Processing Solutions:
1. Low Viscosity Fluids
a. Immersion
(1) Maurer Monobath Processor
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
(2)
Fairchild Minirapid Processor
(3)
Maurer Cine-Instant Processor
b.
Jet
Spray
(1)
Kelvin-Hughes SAGE Processor
(2)
Ansco (Berley) Spray Chamber Processor
c. Cell
or Chamber Applicator
(1)
Ansco KD-11,-12,-13,-14, and 9 by 9 Industrial Camera-
Processor Printer
(2)
Ansco Inflight F at Processing Magazine
(3)
Optomechanisms KD-5
(4)
Specialties Rapromatic 4000
(5)
Aeronutronics Capillary Chamber
(6)
Signal Corps Stepwise Processor
(7)
C.E.C. "Datarite"
(8)
Hycon 9 1/2-inch Rapid Processor
B. Indirectly Applied Solutions
1. Low Viscosity Fluids
a. Saturated Web
(1) Rapromatic, Inc., Raproroll Processor
(2) Chicago Aerial Industries Inflight Processing Magazine
for KA-30 Aerial Camera
(3) Eastman Kodak Bimat Process
(4) Rapid Processing Monobath Cassette, (H.R.B. Singer)
b. Porous Roller or Plate Applicator
(1) Photomechanisms Rapidata Processors - 35 mm and 9 1/2-
inch
(2) EK 5-inch Inflight Processor
c. Nonporous Roller Applicator
(1) USAS-RDL Wheel Processor
2. High Viscosity Fluids
C. Diffusion Transfer Reversal
1. High Viscosity Fluids
a. Polaroid Continuous Strip Processors for Aerial Film - 70 mm,
5-inch, and 9 1/2-inch films
b. Polaroid 35 mm KD-5 type Camera-Processor-Projector
2. Low Viscosity Fluids
D. Unconventional, Non-Silver or Dry Processes
1. Electrostatic Systems
a. Continuous-Tone Xerographic Recorder
b. Haloid Xerox PROX1 Processor-Projector
Table 3-1 lists current processors, and gives their major features.
Originally, Aerial Photography was accomplished through the use of rather
large negative formats, centering about a 9 by 9-inch negative which is still in
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
r
z
C/
w
w~
~0
F..
g
G
--'
CO
QQOQ
G
H
co
w r0.
G G
ro=i
rt
i
~
CD
0
CD
H? C) CD
~
O H
(or,
t
m
0 w.
)
II
C)
H
v
(D
N
F
-?
a
"
C
w
0
n ro
n
H.
ro
ro
ro
?
t
C
0
M n
D
D
C
C
CD ro
C
D
i
CD
CD (
D
C
rt M
Fi P)
CO
CO
CO
CO
CO
C)
rt M
M
M
CO
CA
M
M H.
--d
0
o
O
O
0
:3
Pt
H
I
" (
F
S
I
i
~
t
~J
z
Z
b --~ -C0
-60-
w C)
fA-
-03-
C)
~,
H a 00
rP
H- CD C!;
y
cn
w 9 CO
0 C"
H. rt
CA
0
H
M (D
0
C
r' N
~F--'
r
a
O
~'
0 0
0
'
(DD
M A)
I.-
Fj
rt
~
a
a
a
a
a>
~
~
.o
FAJ
%
FP N2 FP
Z
C) F-'
- ----- -----
F-
------- --
0~
P) CAD
0
M
0
9
0 w
.o
p)
0,
`
0
~
F
d
~C M M
~C)1 'T0J
$
1 O
I- M
Q
" Zi
T
w
m '
?
M
t
(DD rt r
ri
w
l
co
rt
F' 00
CTF
t7 w
F+.
C
rj rt
r-
A
M
0
.
G
w
F-'
9
C
n
V
F
to
0
~cn
0) M
OD
H C+7
=
Cl,
o
`
~C
c0i
a
=7
~t
9
9
H
QQ 9C
I
{
CO
M
V)
M M
rt Cn
F-'
co
F---'
H
N
i
Ct
60
F~00
H H tD
-
(D o 10 F%'
-----------
r?
w H
0 CD N
MM ~~C
P-h
0
to
14."i O (D
M O CL
F-h M CD
0 H? rr
0
H? ro C)
O' CD
0 w
CD - I-+ O
C
Err
n
'
f)
Cwt CD (D M
H rt CD
E H (D
C CL
.. Hl M
R) t F-'
H
fD
p
0
(p
CD I--'
co H.
C d CD O
D
M 0
(D CO
-
0
CL
CD 0
O
N
v
C
r
rFG
h~
(D
a
C
I
C D
rt
f
r
r
N W ?
H. C!) OQ
rrt 1-'
Cl) P. +
0) O W rt
Cl)
CD
0,
0
F-'
H
H.
C
0
(D
o
rt
w fD
G OQ I--'
CD
H?
0 OA
G
00 0
(D 0' H
ri O
M r
!
w
M'4 b
F-' w
rt C].
M rt F{.
OQ CD O Fl-
M f) I-h
F-' w
w M Q?
H? 'd
CD M w
H? H-
o '^C
'd 0
;j a
n.
H? 0 w
rt w
H
rt G
H? o
'
W M
H.
C).
CO) rt H.
H. C)
M C rt
G (IQ
H
rt 'b o
w d 0
1.- QQ
O'
rt Ft
y' rt
H- H. w
0'.
(D M
H b
0 w
C1.rt
w
MId
w
0~-JFj0
14
M
CQ
F
H
Q
m p4
,Q.
0
rt
C)
H? CD o
.
I
-'
C
'd w E
H M
(
D
M 9
0 0
w ht
M H
F-' 0 M
H? Fh
n 0
CD
rt
rt OQ CD
G 14
w roan
~
0 W H? {
M
- 00
(D rt
"
2 H?
w
.4
CO
x+
9
D
0
o
co
A
0
?
rt
?
?
~
D
CD rt (
CO
(
D
C
CL Q
t M
rt
p.
rt
"
'.
C
M
r
t "
0'
F- H
~?
rrt
F
CO H-
r r
rrt
C
(D
' j H M
F
rrt P)
H M
H.
~
M rt H. t-4
w
o'? d
F
G
I-'M
~H H-
r_
rt
H- F-
N? r
t
.' H
7
F
-h 0
;:r
M
M
H?
r
M
M
C a
CD P ID H?
0 +rJ w
{
I- H.
93
- o
? I
-h
M
F-
Q Q C). M
G
'
Fi H? rt M
M rt
M
O
tD
F?
CD F-' H? DC
I
(D
D
H
w
CL
E P.
CC)
S
w
FI.
? w CL
M
I-
9 C' t
{
_ ,'~ M
H
rt o H.
'
rt
H? QQ
H
~
rt l
I
'
M (D
M ft
w
f
..i O G I-h
0
H
C
d CD
C
O C] r
rr
0
w
H- CD
t CL
H. rt
y> r H.
'
H
0 Fh
0
0 P3 1
9
a
N
r
H?
H O
al
C
tf 0. CD
t f H C-
'4
O H.
H Fr
H
I
M
- H. i
H -
D
0.
0 19
M
v Cs.
QQ rr
F?
C
F4
rt
CD
M
?D
C
D
G
G
M rrt O
rtt 0
'
w
(
M
E
A 0 t D
0
P3
CO
P)
I
rt rt pP G
CO) rt
CD
AH
~
O
it
r
tr
'
n
y
(D
1--?O 5C
F
t
W
CD
rt rt (D
M
o
CO
. n
O M
C
"
H- ~
P. 0' CD
(
D
co
? b w
N
w
I
'4
P)
_
HH
CA
0
G
Q
?
CO
CD
Orr
H
(0
r
n
?
M
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
4)
1.1
r?
0
A
-4
W i1
Cd
-
W
11
to
cd
,-a
to
4!
4)
M
to
4)
0
U
0
(
1
.1 a
O
d
)+
N
JJ
Q
0
O
4)
C4
ad .
~
w
C>
4J
Qy
1-4 0
O
4a
()
0
41
it
-H ~4
k
14
-,4
0 H
d z
9: 0
0
0
4)
is
4a
rt
4)
-
.0 0
a
-
.-t
0
.4 4j
'LS if
O
,
pp
0
m
a!
1
?a s
-1
0)
Cd
,
C
t1
0
m?
0
r
Ua
~
.
14
w
0.
0
0
a!11
d+
+
U
.
-r4
4-1
U
0
N
r+ w
4
'0
?+
a
N a
D
;1
~4
t
0 r
A
tti j
O 0
'C
Ci
t
+
41
-4
c)
a) a)
~
>
r-I
0
0
-It
0
0 0
r-1 O
4)
0
44
4) A 4)
,?a ?
,.t
co
,-4
0 9
A
0 0
w 1.1 a
ri
9
0
4)
-rt -rt
11
0.0)
t
-
0
0
w k
u
0
0 r- 44
wea
4
'~
4)
w0
)+
-10)
r
t14 1
>a0
o 14
'zs a)
a
'v ao
0 4)
a 0
'G
?
es
t
1
1
F N
to 0)
$4
4)
m r1
A
Cd
4) 0
0 0 )
t0
U
> )+
O 0)
.
r
0 C1
W
0
`?
-
W
H a)
-1
11 ,~ 0
0 .0
11 P.
4) 'r+
.-
,
U
4)
U U H
U a
4)
H
.-4 -r1
E U
P. T
0 )+
0
4)0 0
4)P
0
0. to
11 .
0. 0
0 1-1
-rl 0
)a 4a
4-1 4) --l
H
P
0
r-1 U
0 'ri
U -r4
I 4.1
0 a
01 11 Cd
0
0
t0
r~ )+
) 'G
CV
b0
as 4)
?a 4)
.0
-rt
9
4) P.
0) rt
\ 4)
,-1
411 ,-d
1.1 't3 1
H
AJ )1'
41 rr
y
tm
4 ( .0 4
0 0 0
0 ad
0 -r4
4) V.
-rt
E
4)
H .l-1 t0
u3 44
4
?, ?
z 'i-t
-,a
i- .D
G'1
A
io
b -
N
N
x
x
N
1-4 H
-4-t
Low
oco
z
mot'r
10 r'
CO -4
z
z
z
r4
H
-4
H
r4
1-1
a x
x
4
0 d
41
4a
-
r
i
z
c)1
N r-1
z
0
n
r1
y
y
,4
4
0 4)
U
to
r-1 -4
to
r-i
,-+ 0
u
0
r-1
CA
ra
'>
r-t
r1
H
'- (}
O
,4
;;r.
41
w
CA
11
c4 Eq
4)
1 O
4 .1
s0 0
~
H
0
(.
3 ~
4)
aA
A.a ;~
4) U
0
E U
d
id
a,
W
O O
O 4S
r-1
O
?
to
-4
\
z
\
z
fsG
r4 (0
r-+ '
-l
L`7 cD
l
,
U)
Ei
R5 i~
W Lz]
E-j
.4 HH
a
z
z
z
z
z
z
z
41
~.
0
0 cc
t
to k
0
4
0
000
CD C> -4
CD U '-4
0
O O 4) ,
0
G
0
0
41
\
CA
O
. 14
to P+ L- 1
.
0
N- '0 0
\
?-.
0
1
U
00 ' v
O
-e~ U
-(1)-
z
z
z
l
z
t
0
O
'4
~4
O
0)
?!
03
4
44
aad
44))
o
P
0
0
P4
O
4)
0
0
$4
W W v
W
1
4 41
P~
4)
4)
1
U
4) 44))
m 0
M rl
i
W H
t 41
ad U
,-1 )a
.--1 4)
0)
0
0 1
.,1 m
1
CO to
0 P
~.
)+ 1.1
43 N
M 4)
9
H 4)
r-1
4)
N (L)
00 0
)-1 00))
U b) O
)+ 4)
gU
iU
O
6
14
}0 0
4
8
.1
8 0
..
LL
U
0
W
P
a
1
[J'
z
OP,
1++ r,
O-
.a
OP
0
4
> En
/
0)
U
U
0
'o
W
0
)1
0
4)
41
-4
-.4
4
4J
v
0
ti
0
t 0
0 4d
9
0 0
,-t
)+
W
0
O
U
"4
ca
41
0
11
U
P4 lot
5.-
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 2
z
v
z
6)
? m P.'d
H m
00
02 m
G "U
? r
cn
rt
H?
n
0OQ
o' m
0
n
m
0)
m
0
H
0
m
0
rt
0
Cn m
rit
rt
rt rmt 0 ~
H? -1
a ~. o0
an m
n
M rt
za0
m m a
m O)
0 4
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
N r w M
C- CCD
H
co
M 'd m H?
aim rt n m + m I wm
f 1 ' ? O
a G q
m
m ti
K w
02 -m
rt ~rt
W
rt
rn
0
O
0
H?
a
m
rr
m
m o
O N H
rt o m
0, co
m rt
03 N 0
n 0 m
rr
b r
rt rA 0
P) m
m t1
x
d m w
H loo
fD rrP
F'- a
~ rt m
h'h Cl)
W 0 rt
r a
~ r
K
O
CD
m
02
m
0
H
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
03
a w
a4
m
z
6)
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
LO
ti
?4 iS
d4
0
N
-4 C%I
tA-
i4
k4
14
x
0
W m
m N
m cG
00 m -O
0 mLOYi
14
m N
4
r4 74
A i4
- X
r4
to
O oo CD "14 00
M H
ri tfJ
CG t[~
N
b- 0 r4 C4
M
4
N
O N W N m .m0 4a0
!t? I-I C) to
N C) OD
m w
1 w w i3
a.4 r4 4 4
a O to
,-4 Cq v N r4 a 0 r, y
0 to
P+ 1 m 7 0
fn
L
O r
o
a r a40. LO r-4
~4 r4 T-4
00 a 1 ,-4 CO r4 U a o
m M a w -a A
w m Cd w r1 w U m co u
m m r+ C a V) u 0 4.3 41 >1
41 4J C6 4J V Aj PN 41 cu
41 41
4.1 r4 CO 1 r-4 4.7 U U 3 11 U
3? D 3~ cCO 0 v 3m C
to
CO 400 0 1 O 1 0 Q'. 0 1 G
O -' N 0O LI0 000 CVO Z LOO0 r
~y ,--t O 9 N 1-4 II) r4 LO N (0
6 d
z z z z z
a
0
O ? ?' 0 O
0
r4 0 o O N
? M 00 O
w to rLO
)
m
O
1-
a
U
0
4.t
?
0
00
v
m
,a
a
,~
m
m
'4
a
u
13
u
v
~4
C:
,4
44
0
?d .-4 r-1
ra N
4v d A m
N
0
U
" u c01
41 0 -4
4 ft it
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
rt
r-I
0
14
a
at
u
tYS ~ ra 41 a. a W
.C
m
P4
IJ
to
U
', V 41
4d
F 00 0 -
"4
v4 $4
w
0
04
00
0
a
rd
0
0
d] ?
i+
41
4:
W
H ?~+
00
a 00
f
O
d j
U
.J:
m w ,
0
I
a
a
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
%0 %W
SPECIAL HANDLING
use today, and frequently including the 9 by 18-inch negative, as well as other
specialized systems encompassing several feet in their maximum dimensions. Al-
though enlargers have been built to handle large format negatives, the bulk of
printing has been by contact. For normal applications the significant detail
will be acceptable.
Modern trends have been toward much smaller negative formats which have
resulted in the development of more precise projection equipment to achieve
optimum results from these smaller negatives. The enlarger is used to secure
a print of given size whether enlarged or reduced. The contact print is by no
means obsolete, but the enlarger, or projection printer, is the primary means
for conveniently producing prints.
Improvements in the ability of modern aerial reconnaissance systems to re-
solve greater detail have necessitated comparative improvements in enlarging
equipment. These trends have been toward instruments which can, at a rapid
rate, project images onto film or paper, control the contrast of the finished
print, and rectify any distortions apparent in the original photography. For
slower or more conventional print production rates, the step-and-repeat printer
(a single exposure at a time) is usually used, while for higher production out-
put the continuous printers are employed.
The high speed continuous printer is basically a unit which will accept
long lengths of negative input, as well as an equal length of unexposed film or
paper. The two are sandwiched under tension across a drum in which a slit,
above a light source, provides the actual exposure. These units do provide a
degree of exposure control through such means as varying slit widths and lamp
intensities. At least one continuous printer is in use which provides an
electronic contrast control.
3.5 AUTOMATIC CONTRAST CONTROL (Dodging)
Pictorial photographers have for a long time tried to compensate for the
restricted brightness range of a paper print. The loss in tone quality, as well
as information, has led to various printing tricks, from simple shading and
local overprinting during the enlarging, to more advanced methods of deliberate
distortion.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
What in a pictorial photograph is merely a lack of aesthetic quality be-
comes a serious matter of information content in specialized applications such
as aerial photography. Trial and error methods of contrast control in such
cases are inefficient, expensive, and unpredictable.
This section provides a brief description of the various approaches to
contrast control. It is assumed that the reader is aware of the general use of
various contrast grades of photographic paper. The difference between contrast
control through the use of varying contrast grades, as compared to the various
other employable techniques, which not only reduce contrast but actually gain in
detail, is most important.
The use of various printing paper grades for contrast control can somewhat
compress a negative tone range to that of a printing paper, but only at the ex-
pense of detail contrast, and usually with a notable loss of detail near the
ends of the tone range. The masking processes, widely used in the graphic arts
field, get over some of these difficulties.
The obvious difference between the use of varying paper grades, and the
introduction of masking of any sort, is basically that through the use of the
mask modulation of the printing light has occurred. All efficient contrast
control systems depend upon control of this modulation in one form or another.
The following variations will be discussed:
1. Variable contrast papers
2. Hand dodging
3. Manual dodging printers
4. Unsharp mask techniques
5. The electronic methods
a. Velocity modulation
b. Intensity modulation
6. Quenchable phosphors
7. Photochromic or phototropic masking films
Included with No. 6 is a brief description as to its ability to produce effec-
tive edge enhancement with no additional laboratory equipment.
3.5.1 Variable Contrast Papers
For all practical purposes, the contrast or printing capacity of a paper
may be defined in terms of its useful exposure scale, or inherent gradation.
3-24 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Photographic papers in which this image contrast may be varied over a wide range
by changes in the color of the exposing light have been introduced in recent
years. With these papers it is possible, by use of filters, to duplicate on one
paper, the range of contrast obtained on the different contrast grades of other
papers.
There are several distinct methods of achieving this, all of which are
incorporated into the manufacture of the various emulsions.
1. Multigrade is a blend of two emulsions, one of low contrast sensitive
only to blue and violet light; the other of high contrast sensitive to green as
well as to blue and violet light.
2. Varigam is a single emulsion to which a green sensitizer has been
added. It is believed that the sensitizer is absorbed to the larger grains in
greater amounts than to the smaller, with the result that the larger grains are
made more sensitive to green. Thus, with an emulsion in which the contrast
varies with the average size of the silver halide grains utilized in the expo-
sure, the use of a yellow filter will confine the exposure to the larger grains,
producing a soft print, while a blue filter will cause only the smaller grains
to be affected, producing a print of higher contrast. If a single exposure is
made through a filter transmitting both blue and green, an intermediate degree
of contrast is obtained, depending upon the proportion of each transmitted.
3. "Unicontrast" type is not true variable contrast paper, but simply
emulsions with a characteristic curve of such shape that the contrast of the
print may be adjusted to the requirements of the negative by exposing so as to
use the proper part of the paper curve.
When printing with variable contrast papers, it is possible, and often
desirable, to vary the contrast in separate areas of the photograph. By using
different filters to expose separate areas of the print, a harshly-lighted area
can be softened, or a thin flat area snapped up, without changing the contrast
of the rest of the print. This gives the effect of several grades of contrast
in a single sheet of paper.
It is important to realize that the major difference between the use of
varying paper grades to achieve contrast control, as compared to the more exotic
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
-,
SPECIAL HANDLING
methods such as the electronic or quenchable phosphor, is that paper grade
variation subdues gross detail and reduces the gross contrast, while the elec-
tronic and quenchable phosphor systems maintain the detail contrast. To state
it more simply, the electronic or quenchable phosphor control compresses the
overall tone range but keeps the detail contrast at any chosen value. In prac-
tice both these principles are capable of compression ratios far in excess of
any known paper grade.
3.5.2 Hand Dodging
Exposure is varied between relatively large areas by a hand-held mask
placed between the printing light and the sensitized material being exposed.
Although judgment is permitted, this is a laborious and time-consuming task.
Small area contrast control is not obtained.
3.5.3 Manual Dodging Printers
Many printers presently in use make provision for manual dodging during
exposure by using as the exposure light source a large number of low-power,
mercury-vapor type lamps which can individually be extinguished by manually
throwing switches associated with the lamps. The operator can, therefore, view
the negatives and adjust the light intensities of the individual lamps compris-
ing the light source in accordance with variations in the density of the nega-
tive to compensate within limits for the varying undesirable densities in the
negative. Such manual manipulation to satisfactorily dodge the negatives, how-
ever, requires long experience and good judgment, and can be effectively prac-
ticed by only a few highly skilled operators. Also, because of the very careful
inspection of the minute increments making up the total areas of each negative
which is required to effect proper adjustment of the bank of lamps forming the
exposure source, this practice is both time-consuming and laborious. Printers
of this description commonly accept negatives of formats up to 10 by 10 inches,
while larger printers accepting formats up to 10 by 20 inches are also in every-
day use, and are essentially a one-man bench-top operation. Because of the set-
up time involved, they are not normally considered high-production printing
equipment.
3-26 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
3.5.4 Unsharp Mask Techniques
Light modulation is accomplished by placing a previously-prepared unsharp
positive of the negative being printed between the printing light and the nega-
tive. In this process, light modulation is applied to somewhat smaller areas
than in hand dodging. The system is time consuming because of the need for
careful registry between the unsharp positive and the negative.
The overall tone compression is still not selective; it affects the whole
image. To preserve both shadow and highlight contrast requires multiple mask
systems, leading to various laborious tone separation processes based on exten-
sive densitometric evaluation of the original negative.
3.5.5 Electronic Control
This method has been successfully employed in both contact and projection
printing systems, and until quite recently was the most widely used contrast
control method employed by the military and large civilian laboratories. The
electronic dodging printers are manufactured by only two companies at this date,
namely, LogEtronic of Alexandria, Virginia, and Rank-Cintel of England. Each is
similar in that both employ a scanning light beam from a cathode-ray tube as the
printing light source. Two methods by which this light may be modulated are:
(1) intensity modulation, and (2) velocity modulation.
In intensity modulation, the scanning beam passes over the different parts
of the negative beneath which a photo-multiplier tube monitors the light coming
through. When scanning a dense negative area, the light reaching the photo-
multiplier tube is reduced; when the beam passes over thin areas, more light
gets through. An inverse feedback amplifier receives the output from the photo-
multiplier and, in turn, controls the intensity of the scanning spot in the
cathode-ray tube. The reduced light from dense negative areas thus makes the
scanning spot more intense to compensate; when the beam passes over a thin por-
tion of the negative image, the feedback circuit reduces its intensity.
Velocity modulation differs somewhat in that the photosensor provides a
feedback signal to the cathode-ray tube, which causes a simultaneous and con-
trolled variation in the speed of the moving spot on the face of the tube. The
scanning speed at any given instant is inversely proportional to the density of
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
the negative in the area of the spot. Automatic contrast control is thus pro-
vided in areas as small as the moving spot. With either of these methods, the
net effect is that the negative is printed with a light source that corresponds
to a luminous positive image. The printing light is thus modulated to suit the
negative instead of being continuous. This "light mask" is formed automatically
during the printing process and so requires no laborious preparations or proc-
essing. It is automatically related to the negative, and is in register with it,
and the mask density and contrast are controllable by adjusting the feedback of
the amplifier. However, this type of printer cannot compensate for brightness
variations over areas smaller than the scanning spot.
In practice, a negative of high contrast, but with comparatively little
fine detail, would be printed with a fairly large (3/8 inch approximately) spot
size and appreciable feedback. A negative of lower overall contrast, but con-
taining a great deal of fine detail (aerial photography) would need a small spot
size with moderate dodging.
An objection frequently voiced against the electronic dodging system is its
inherent ability to produce "fringed" edges. As the flying spot scans a boundary
of the image from light to dark, the light incident on the photomultiplier
changes gradually. When the spot straddles such a boundary, the shadow area
begins to relieve more exposure than the shadow region just scanned by the spot.
The edge of the highlight gets less exposure. The result is a dark fringe
within the border of the shadow area, and a light fringe in the border of the
highlight. With a fixed scanning speed, this "fringing" effect is greatest
with a large spot and least with a small one.
The electronic means resorted to for effecting automatic dodging of aerial
photographs are highly complex and have, therefore, produce maintenance problems
of an extensive nature. However, when operated by skilled technicians, the re-
sults are outstanding.
To illustrate the flexibility of electronic printing, the LogEtronic En-
larger is claimed to handle a negative density range up to 2.2. This can be
compressed to a range of 0.3 on the print; in terms of actual contrast, this is
a compression factor of about 80 to 1. The claim for the Cintel equipment spec-
ifies a maximum density range of 3.0 and a print limit of 0.1.
3-28 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
%W %of
SPECIAL HANDLING
The electronic units are large, console type averaging approximately 25 by
40 by 68 inches high. These require factory installation, with special instruc-
tion by factory representative and are quite expensive. Operation is much more
involved than convention printing, but basically it is still a one-man operation
with the possible addition of another man for maintenance assistance.
3.5.6 Quenchable Phosphors
A printing method in which a phosphor plate is made to glow brightly when
energized by ultraviolet light, but becomes quenched upon the application of
sufficient infrared light. Since normal printing emulsions are not sensitive
to infrared, it is possible to pass this quenching energy right through the sup-
port and the emulsion of the unexposed material, through the original negative
(where the infrared is modulated by the imagery), and on to quench the phosphor
plate, almost entirely under very thin areas of the original, and little or none
under the most dense areas. The differentially quenched phosphor plate thus be-
comes the "unsharp positive" which is automatically in perfect register with
the original. This unsharp positive is also the light source which exposes the
sensitive emulsions of the film or paper to produce positive "dodged" copy. In
an enlarging system utilizing this principle, the phosphor and the negative are
in quite close contact. The infrared source is directed toward the negative but
outside the field of the lens. The exposure is similar to that made in contact;
the infrared modulating, through the negative, the blue activated phosphor. A
great deal of contrast control is available through variations in voltage on
both the ultraviolet and infrared power sources. The quenchable phosphor method
of contrast control presents many advantages over others in that its actual
dodging increment is infinitely smaller. This area can well be defined as the
size of the actual grains that make up the phosphor, proportional to their sep-
aration from the negative. This, in effect, means an infinite number of light
sources for incremental dodging, with automatic control of the intensity of each
minute source by the ability to use the negative as the modulating medium.
The quenchable phosphor printers also offer, through use of a special
screen, a very effective method of edge (or image) enhancement. This is accom-
plished by taking advantage of the slight halation effect produced on image
boundaries by the glowing phosphor. In effect, this actually enhances the
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
difference between boundaries and can actually be carried to the point where
only the edges are imaged. This capability is useful for emphasizing man-made
objects for the photointerpreter, making hasty "maps" from aerial photographs,
or eliminating the need for halftone resolution losses in preparing plates for
reproduction.
The simplicity of operation and low maintenance costs of these printers
are most important considerations. At this date both contact and projection
printers are available from the Watson Electronic & Engineering Co., Alexandria,
Virginia; and from the Kelsh Instrument Co., Baltimore, Maryland. The Kelsh
Instrument is a projection unit utilizing optics for essentially a 1:1 output.
The density compression through use of quenchable phosphors is (at this
writing) not as high as that which can be obtained electronically. Neither
printer will print through densities much in excess of 2.0. The Watson Printers
will compress a negative density range of 0.25 - 1.80 to 0.60 - 0.80. The
Kelsh will compress a negative density range of 0.3 - 1.5 to 0.6 - 0.9.
All these printers operate on conventional 115 volt, 60 cycle alternating
current; they are compact, bench-top operations, require no special skills or
maintenance, and can be operated by one man.
The Watson (Fluoro-Dodge) printers average approximately 17 by 22 by 35
inches including reels. Its weight is approximately 85 pounds.
3.5.7 Photochromic or Phototropic Masking
This approach to dodging, or contrast control, is quite similar to the
method of unsharp masking described at the start of this survey. The main dif-
ference is that the mask material is not a silver halide emulsion, but rather
an ultraviolet sensitive dye in a clear film base. This material is exposed to
ultraviolet light while in contact with the original negative. No processing is
required. The two are then printed as a sandwich. The photochromic densities
are proportionate to the densities in the negative; hence the brightness range
is compressed and considerably more detail is picked up in areas which would
otherwise yield none. Following the printing operation, the photochromic
material can be erased by means of heat and used with another negative.
3-30 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
At this date the photochromic mask, though workable, is not truly neutral
but displays rather a magenta hue which somewhat limits its applications with
certain films or papers; however, it is being steadily improved. The use of
photochromics as a dodging medium would necessitate the addition of ultraviolet
light systems as well as a heater for erasure to conventional enlargers. This
system has merit in that it would be by far the cheapest and easiest to adapt
to existing printing equipment.
The production rates on any of the three systems described are completely
compatible with any others in use today with the possible exception of high-
speed xenon flash projection printers. If specialized controls are to be ap-
plied, or a high degree of resolution is to be attained, an awareness of some
of the limitations must be forthcoming.
One of the most important aspects of the data handling problem is the
duplication process. In this section, the photographic aspects of the duplica-
tion process that can affect the quality of the final product are discussed,
and not the specific equipment (i.e., printers, processors, enlargers, etc.).
There are two main aspects of the photographic image than can be altered by the
duplication process; namely, tone reproduction and image quality.
3.6.1 Tone Reproduction
Tone reproduction is that aspect of photographic science which evaluates
the relationship between the duplication process and the original object. The
correct utilization of tone reproduction theory can greatly aid in the evalua-
tion of the proper process gammas to determine the optimum reproduction tech-
nique. In fact, for precision photographic processing and the derivation of
the maximum amount of information from the photo-optical camera system, the use
of tone reproduction is mandatory. The following discussion will serve to
illustrate the need and use of tone reproduction in aerial photography.
Classical tone reproduction theory states that if the log object luminance
range of the scene is recorded on the straight line portion of the D-log-E
curve, a linear reproduction will be achieved. For typical ground photography,
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
this is true. However, for aerial photography, this is generally not true.
For example, the greatest tonal degrading factor in aerial photography is the
intervening atmosphere since the atmosphere adds a non-linear degradation to
the ground scene object contrast. This non-linear degradation of the ground
scene contrasts is not corrected for by recording on the straight line portion
of either the D-log-E curve of the duplicating or original taking film.
In aerial photography, several factors can act to alter the tone reproduc-
tion of the final product. Aside from the atmosphere, there is camera flare,
the characteristic curves of the duplicating stock, printer flare, the viewer
flare, etc. However to illustrate the point, the discussion will continue
with the atmospheric effect and the application of tone reproduction theory.
As stated, the atmosphere complicates the photorecording process, and it does
so in two ways: (1) it reduces the object luminance through scatter and ab-
sorption and (2) it increases object luminance through the addition of its own
energy. This phenomenon can be expressed mathematically as:
B o = I s R g T a + B a,
where Bo = luminance of the object above the atmosphere,
I s = solar illumination,
Rg = object reflectance;,
Ta = atmospheric transmission, and
Ba = atmospheric luminance.
If there were no atmosphere, and no camera flare, there would be a linear
relationship between camera exposure and ground object luminance. Because of
the atmosphere, the relationship is non-linear. (Camera flare will be discussed
later.) A simple relationship can be established to calculate the effect of
atmospheric luminance on a given camera system.
On-axis light losses through a lens can be expressed as:
3-32 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
Bo T1
If = ,
4( f/no. )2 F
where if = illumination in focal plane,
Ti = lens transmission,
f/no. = relative aperture,
F = filter factor, and
Bo = object luminance.
(3.2)
Substituting Eq. (3.1) in Eq. (3.2), and taking camera exposure time into
account, Eq. (3.3) is obtained.
t(Is Rg Ta + Ba) T1
4(f/no.)2F
where E = exposure in foot-candle seconds,
t = exposure time,
F = filter factor,
f/no. = relative aperture,
Is = solar illumination,
Rg = object reflectance,
Ta = atmospheric transmission,
Ba = atmospheric luminance, and
T1 = lens transmission.
(3.3)
Eq. (3.3) can be made to yield exposure in meter-candle-seconds, a term more
commonly employed in film evaluation by the multiplication of the numerator
by 10.76 (1 foot-candle = 10.76 meter-candles). Thus, Eq. (3.3) becomes:
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
2.7t(I s Rg Ta + Ba) Ti
(3.4)
(f/no. )2F
Eq. (3.4) can now be used to calculate the effect of a given atmosphere on the
ground scene. Fig. 3-3 shows how this effect alters the camera exposure. For
a given range of ground luminance, the effect of a constant atmosphere was cal-
culated. The assumed lens and atmospheric conditions are noted in the figure.
From this figure, the exposure resulting after atmospheric attenuation is clear-
ly demonstrated. The contrast loss accrued by the atmosphere is illustrated by
object b and its background b', which are presented to the film as a and a' with
a definite loss in contrast.
The figure further emphasizes an important point that the atmospheric at-
tenuation, being uniform, produces a non-linear alteration of object luminances.
This non-linear degradation of the original scene complicates the recording
process in that a simple increase in gamma, to increase the recorded object con-
trast, does not truly compensate for the alteration in original object contrast.
The purpose of the reconnaissance mission is to record information about the
ground scene and, ultimately, the greatest information will be recorded when the
ground scene is perfectly reproduced. Obtaining a perfect reproduction is, how-
ever, difficult unless one considers not only the taking material, but the dup-
licating material as well. For optimum reproduction of ground scene detail
(from the tone reproduction standpoint), it is necessary to select both the
proper negative and positive duplicating films.
To aid in the selection of the appropriate films (both taking and duplicat-
ing) for a given mission, tone reproduction theory can gainfully be employed.
Fig. 3-4 shows a typical tone reproduction cycle in aerial photography. The
original luminance is altered by the atmosphere and, as shown in the figure,
presents the film with an altered camera exposure. This camera exposure be-
comes the exposure for the negative material. When processed, the density of
the camera negative images become the log exposure for the positive stage. If
the first generation positive is the last stage in the reproduction process,
then the reproduction can be characterized by a plot of density versus log
luminance of the original ground objects.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
11
CID
boo N
II \
II
U
II
Q)
L
"
LO
d er
,
,
O p1
II
En O
w
H
a
w
I
I
.I
cd
I
cd
ao CO eM
I_ Iri I~
a.xnsodxa v.xam 3 2oZ
I,
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
..JH
.- - .. 0 G 0
S719uaU ati11190d
SPECIAL HANDLING
r-4 cd ?r1
co P k
Cl O
?rl 'b
4-4
0
0
d+ C U
I H a
0 0
4J P
coo a
w a. P
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Fig. 3-4 illustrates photography on EK type 4404 film, and duplicated on
8430. The examination of the figure demonstrates the imperfection of the final
reproduction. Although the shadow detail has been somewhat enhanced, it is still
recorded at a contrast lower than the original contrast, while the highlights
have been degraded due to the limitations in the photographic materials.
Consider the case of extremely clear weather. Fig. 3-5 illustrates a case
of photography taken on SO-190, and duplicated on 8430. In this case the repro-
duction is far worse than that illustrated for the moderate haze condition. The
high contrast of the SO-190 has recorded the scene contrast at a density range
greater than the duplicating film can accommodate. As a result, the highlight
information and some medium tone information has been seriously degraded and dis-
torted during the reproduction process. This example could have been signifi-
cantly improved by processing the original photography at a low gamma instead of
the typical high gamma. Fig. 3-6 illustrates the same clear weather condition,
but the substitution of a low gamma processed negative instead of the high gamma
SO-190. The figure shows that the final reproduction has been considerably im-
proved over the high gamma case.
Fig. 3-6 also illustrates the other use of tone reproduction theory, that
is, the determination of the proper film processing parameters. If the charac-
teristic curve of the taking film and the desired final reproduction is known, it
is possible to determine the optimum duplicating characteristic curve. Such an
example is given in Fig. 3-6 where it was assumed that a perfect reproduction was
desired. The desired positive characteristic curve to obtain a perfect reproduc-
tion with the given negative curve and haze conditions is shown. For these con-
ditions, the desired characteristic positive curves to light, moderate, and heavy
haze are shown in Fig. 3-7. Such curves are difficult to obtain with standard
films and processing conditions. The point is, however, that using tone repro-
duction theory, one can obtain the best compromise between the original negative
characteristics and the duplicating film sensitometry to obtain the best repro-
duction possible with existing materials. Without the use of such theory, it is
possible to lose a considerable amount of information simply through the use of
the wrong characteristic curves in the taking and duplication process.
For a variety of haze conditions then, the desired duplicating film charac-
teristic curve could be determined. The above discussion illustrates two points.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
fL
p bc~ Y
Y
Y
i~
1
Y n F Y
~.E { 1
_m
?
?
I
I
I
I
?
I
?
?
?
I s
'
u
n
0.
Y
=
F
3-38
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
M
I/1
ti
~ o
N
U
f
q
o9
y y ?~
?y^ : II
II N
II in
C71
H N '1~1y v ?
v
g uS la V 4 H 6 8 ~
z P,
4
ti N M y~
^ - - - O 0 0 C
6;j U (J OAVITOod
asnsodxg ziamj 8o1
t- M O L- N M
w O O O O O I.+ ?+ I r+ I-
SPECIAL HANDLING
3-39
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
i
Heavy
Moderate
r
-Light
0 1.0 2.0
Fig. 3-7 - Desired characteristic curves of positive duping film for perfect
reproduction of ground scene tonal relationships for three haze conditions
and 4404 negative material as shown in Fig. 6 -7
3-40 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W Vfte
SPECIAL HANDLING
First, simple contrast enhancement through increased gamma does not of itself
correct for the distortions introduced in the photorecording process by haze or
flare. Second, for maximum information transfer and optimum tone reproduction,
the selection of a camera negative film cannot be made independently of the
duplication process, and vice versa.
3.6.2 Image Quality in Duplication
Of equal importance with tone reproduction in the duplication process, is
image quality. Image quality can be affected at two stages - processing and
printing.
Processing - In the processing station, there is very little that can be
done to drastically alter the quality of the recorded image, short of gross over
or under processing. Most of the serious effects on image quality in the proc-
essing station relates to physical damage. The most common damage is scratching,
breakage, physical damage to the emulsion due to too high a temperature in either
the wet or dry stages, reticulation due to temperature or chemical pH, etc.
There are less severe, but nonetheless important physical changes that can occur
during processing. The importance of these depends on the film usage. Dimen-
sional changes during processing can be a problem if the film is to be used in
taking accurate measurements. There are two types of processing size changes -
general and local. The general processing size changes are primarily related to
the physical processing problem. There are several reasons why processing di-
mensional changes take place and the reasons differ for cellulose, triacitate
and polyester base films. In the cellulose triacitate films, a shrinkage occurs
due to the loss of solvent from the base or to mechanical strain release of the
gelatin. Typical values for processing dimensional change in the cellulose
triacitate base films is generally -0.06 to -0.10 percent under reasonable proc-
essing conditions. Processing dimensional changes with the polyester base films
are solely the results of mechanical effects of the gelatin layers. Because of
this fact, the moisture history before and after processing will largely deter-
mine the observed processing size change. Typical size changes for polyester
are still lower than for the triacetate being in the order of 0.01 to 0.03 per-
cent over a reasonable processing range. These effects are general and change
the entire format uniformally.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
There are other effects which are random. These relate to the drying stage
and are caused primarily by drying film with water drops on the emulsion surface.
When water drops are present on the surface of the emulsion, the film will dry at
a different rate. This means that one will be dry before another. The result of
this is a random distribution of image positioning within the format. The amount
of such a distortion is impossible to measure or know. However, it can be sig-
nificant enough to warrant the use of a good air knife or squeeze roller before
the film enters the dry cabinet.
Although processing chemistry generally does not drastically alter the image
quality of the processed film, there are some effects which do occur. Changes in
processing chemistry can have a small effect on image quality. Fig. 3-8 demon-
strates the difference in the transfer function of Plus-X Aerographic (5401) in
two different developers, D-76 and D-19; the D-76 curve being higher than the
D-19 curve. It also can be demonstrated that D-76 developer also produces a
finer granularity than does D-19. Table 3-2 illustrates the granularity compar-
ison for several films.
A similar comparison can be demonstrated in terms of resolving power. Table
3-3 shows the results of resolving power determination on EK type 4404 emulsion
for different chemical formulations.
Although there are some improvements in going to a "fine grain" developer
such as D-76, the magnitude of the improvements is usually small in comparison
to other factors which hamper the image quality of a photographic system. In
fact, there are many situations when going to a fine grain developer can hurt
image quality more than help it and that is through a deleterious effect on film
speed.
In many systems, it is desirable to use a fast shutter time to minimize the
effects of image motion and/or vibrations. The selection of processing chemistry
now becomes extremely important to system image quality if maximum emulsion
speed must be obtained. Although fine grain developers aid film image quality,
they also retard emulsion speed. The use of such developers, when maximum speed
is desired, could well impair image quality if it was necessary to use a longer
exposure time. Hence, film speed is an important parameter in the effect of
processing on image quality.
3-42 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
f-N
D-19
/INS
D-76
2
0 3
0 4
0 5
0 6
0 70 80 90 10
0
Spatial Frequency, cycles per millimeter
Fig. 3-8 - Modulation transfer function of Plus-X Aerographic
(5401) in developers D-19 and D-76 (EK data).
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67BOO657ROO0300210001-4
SPECIAL HANDLING
Table 3-2. Granularity Compari-son for Several Films
Granularity (JD)'
Film
D-19
D-76
Plus-X Aerocon (8401)
0.085
0.058
Plus-X Aerographic (5401)
0.088
0.074
SO-243
0.016
0.013
Aerographic Duplicating (5422)
0.043
0.037
Fine Grain Aerial Duplicating (8430)
0.021
0.020
Table 3-3. Resolving Power Determination on EK Type 4404 Film for
Different Chemical Formulations
Resolving Power in 1/mm
Developer
1/mm
% Diff.
from D-19
1/mm
% Diff.
from D-19
D-19
839
0
278
0
D-19 + D-76
(1:1)
832
-1
295
+6
HC-110
714
-14
269
-3
12-DX-90
684
-17
270
-3
3-44 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67BOO657ROO0300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
The speed of the aerial emulsion is, within reasonable limits, a variable
parameter that can be partially controlled through processing. For example, the
speed of all aerial films can generally be varied over a range of 6:1 or better.
Fig. 3-9 illustrates the speed increase over normal D-19 processing through the
use of a special high energy developer (MX-577). As shown, the speed has been
increased by a factor of approximately 3.3. Under processing, in relation to
the D-19 curve, could reduce the speed by the same factor.
For significant emulsion speed increases, using aerial emulsions, a higher
than normal fog level must be accepted. The amount of fog buildup that is tol-
erable depends upon the system and its intended use. As a general rule, a gross
fog level of 0.4 is usually acceptable for most reconnaissance applications.
This value is selected since with most of the aerial emulsions the beginning and
range of the linear portion of the D-log-E curve is not significantly altered at
0.4 or lower fog levels. For cases where maximum speed and efficiency are de-
sired from an aerial emulsion, exhaustive testing is required. Both the speed
and the developer efficiency (i.e., the speed produced in relation to the resul-
tant fog level) are directly a function of the developer formulation, tempera-
ture, agitation, and development time. The same speed and fog relationship is
not possible with different developer formulations, and some are considerably
more efficient than others. For example, the comparison shown in Fig. 3-9 shows
the MX-577 developer producing an increase by a factor of 3.3 in speed, with a
fog level of 0.3. Such a speed is not obtainable with D-19 with the same fog
level. In fact, such a speed is not obtainable with D-19 under any processing
condition.
The important aspect of this problem is that for any film-developer combin-
ation there is an optimum set of processing conditions, and for the utmost speed
performance from an aerial emulsion these conditions should be determined.
Fig. 3-10 illustrates the speed and fog relationship of S0-243 in D-19 for
several process temperatures. From this plot, it is obvious that 100?F would
produce the optimum efficiency for developing S0-243 in D-19. Maximum speed in
conjunction with lower fog levels are produced at this temperature. For most
practical processing conditions, then, the emulsion speed produced by a given
processor, chemistry, time and temperature, has a more important effect on image
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
0 3.88 Absolute Lo E 1.0
t0 3.88 Absolute Log E 1.0
Relative Log Exposure
Fig. 3-9 - Characteristic curves for EK type 4404 emulsion, showing
increase in film speed possible with high energy developer MX-577
(special EK developer) as compared with normal D-19 processing.
SPECIAL HANDLING
3 min MX-577
at 68?F
/6t min, D-19
68 ?F
2.
0 3.0
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
I
wwwww
00 O L O O
to 00000A O
?%
~
~
1
0
?- _
?
N
? O
,
0 0 0
\.i
r
9
a
i
I
?+
000
I
0
M
O O
SPECIAL HANDLING
W N
0
cq en
w
>
4
O
N
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 +~ 1wl-
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
quality than does the simple effect of developer formulation on the physical film
quality parameters.
Printing - The printing operation can have several effects on photographic
image quality, some of these relate directly to the printer and its character-
istics, others are independent of printer characteristics.
Contact printing of high definition images for maximum image quality re-
quires consideration of a number of factors, all of which influence the formation
of the printed image. When the negative image has low contrast and the printed
image must likewise have low contrast, additional contributing criteria must be
carefully weighed.
Implicit in obtaining maximum image quality with contact printing in gen-
erally a parallel light source, elimination or reduction of stray scattered light
in both the negative and the raw stock, the elimination or reduction of halation
in the printing stock, and the elimination or reduction of interference fringes.
If perfectly parallel illumination could be ensured between the negative
and printing stock, the two materials could be separated indefinitely. However,
in practice this is not possible, so that optimum results are achieved by having
the emulsions of the two materials in direct contact. Any deviation of the light
rays is minimized, since they have the shortest possible distance to travel.
Absolute contact between the two emulsions is made particularly difficult
because the negative image has associated with it a relief image, and because in
reality two surfaces can never be placed directly in contact. However, several
techniques have been used to successfully print resolving power test patterns up
to 300 lines per millimeter.
One technique employs a vacuum to provide differential pressure sufficient
to force the two emulsion surfaces close together over a 10-inch square area.
Unlike conventional vacuum printing frames in which the two materials are sand-
wiched between a flexible membrane and a glass plate, this system compresses the
two materials between a glass plate and a resilient rubber pad. This technique
is useful primarily for step-and-repeat operations, but adaptations may prove
useful for continuous printing.
3-48 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
NW W
SPECIAL HANDLING
A second, and more conventional approach, is to press the two surfaces
together by placing one on a resilient pad, placing the other material over it,
and then forcing them into contact by the application of weighted glass. In-
verting the system has been successfully used on continuous printers.
Resilient or flexible backing for at least one material is necessary to per-
mit the printing emulsion to accommodate to the uneven surface of the negative.
For low contrast negatives, the relief image is not so pronounced, and a more
rigid backing material can be used to advantage.
Conventionally processed negative images have a relief image associated with
the silver image. The height of this relief can be varied to some extent by the
choice of the proper developer. However, even when processed to produce the
least pronounced relief, all parts of the nonuniform surface cannot be placed in
direct contact with the recording material. For this reason, liquid printing is
especially useful in effecting optical contact for eliminating interference pat-
terns and for transferring maximum information from the negative to the raw film.
All the light which strikes the negative image does not proceed directly
through the negative; some is scattered by the negative image, some is scattered
by scratches and striations in the base, some is reflected back from the emul-
sion-base interface, and some is reflected back from the emulsion and air inter-
face. That portion which is reflected does not produce a latent image in the re-
cording material until it has again been reflected back from another interface or
by a clump of silver grains. Unfortunately, these light rays do not produce a
well-defined image, and some degradation is produced because of their effect.
Even the rays which are not reflected from interfaces but are scattered by the
silver image and base-side scratches also produce a loss in image quality.
Shadowgraphs of interference patterns are also a frequent cause of image
tone distortion. Since this phenomenon is caused by a nonuniform spacing between
two surfaces in physical contact, an obvious solution is to have the surfaces in
optical contact or to separate them entirely. The latter solution, however, is
inconsistent with the conditions required for optimum image definition. A more
obvious solution is to establish optical contact between the two emulsion layers.
Even with high vacuum in a step-and-repeat operation, this technique has not
proven satisfactory nor is it expected to be perfectly achieved with dry
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
materials. However, the use of a liquid between the two emulsions overcomes this
difficulty. The liquid must have approximately the same index of refraction as
the emulsion layers, must not cause physical distortion of the emulsion layers,
must be readily removable, and must be nonviscous to permit the two emulsions to
come into close contact. A variety of such liquids have been examined and a num-
ber have been used successfully in such printers. Liquid gate printing, as this
technique is called, is being used in a number of commercial laboratories and ap-
pears to be the most logical way to overcome this difficulty. Finally, some light
which penetrates the recording emulsion is reflected back into the emulsion layer
from the emulsion and base or base and air interfaces. For this reason, the raw
film emulsion is frequently dyed to limit light penetration, and/or the base con-
tains an antihalation dye.
The technique of liquid printing is helpful in restricting the scattering
and reflecting of light between the two emulsion layers, from base-side scratches,
or at the emulsion and air interfaces. However, this technique must be carefully
handled; the liquid must be maintained in a high degree of cleanliness. To be
most effective, the liquid must not be allowed to introduce bubbles, and turbu-
lence must be avoided.
It should be pointed out that parallel illumination does not always produce
the "best" image quality. An example is when EK type 4404 film has imbeded in
the Polloid backing extremely small quartz particles which give the back a
roughness to prevent ferro-typing while spooled on the roll. When printing onto
a high resolution S0-105 duplicating film with parallel light, these particles
print out. The images of these particles is deleterious to the image quality.
In this case, the use of a diffuse source materially improved the image quality
of the final duplicate.
Even assuming that the printer is very good, with a minimum of image quality
loss, there still will be losses in image quality due to the physics of the proc-
ess. For example, one can see that the simple process of making several gener-
ations of duplicates will increase the noise of a duplicate. Film granularity
is a statistical parameter. Thus, since the total variance of a system is the
sum of the individual variances, the sum granularity (aD) will be equal to the
square root of the sum of the individual (o')2. That is:
3-50 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
%00 Afte
SPECIAL HANDLING
2 2 2
6Dtotal - 6Di + aDN
Table 3-4 illustrates the two cases. In case 1, the original photography
was gathered on Pan-X Aerial (EK type 4400 film) and duplicated on 8430. The
granularity of 1, 2, and 3 generations of 8430 duplicates was calculated. It is
seen that the granularity increases as the number of the generation increases,
even though the same film is being used for the duplication. The second example
is for taking on High Resolution Aerial (EK type 4404 film) and printing on 8430.
The same point holds here, as the number of generations increases, the granular-
ity increases. In this case the granularity of the third generation 8430 dupli-
cate is nearly twice the original granularity of EK type 4404 film (0.023), and
almost exactly twice the inherent granularity of 8430 (0.021).
3.6.3 Reversal Processes
Historically, in aerial photography, the positive prints (or transparencies)
used for photointerpretation have been made from a duplicate negative. This neg-
ative is presented from a positive transparency that, in turn, has been printed
from the original negative. Thus, the valuable original negative is used only
once in the duplicating process, and is not subjected to wear and tear through
successive handling. As discussed, however, each stage of the duplication proc-
ess is accompanied by a certain loss in image quality and degradation in tone.
The use of a direct reversal process would seem to offer advantages, therefore,
from the standpoint of reducing the number of generations necessary.
Although abbreviated reversal processes have been reported (see H. K. Howell,
Photo Engineer 5:182, 1952), the most common direct reversal processing technique
is bleach reversal. A bleach reversal process is so named because, after the in-
itial negative is developed, a chemical bleach is used to remove the negative
silver image. This action leaves a positive image of underdeveloped silver salts
which are then exposed to light and developed. When carried out in its most com-
plete form, this process involves the following steps: imagewise exposure, first
development (produces negative image), rinse, bleach, clear, rinse, uniform re-
exposure, second development (produces positive image), fix, wash and dry. Even
though the bleach reversal technique eliminates a printing stage, it possesses
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Table 3-4. Granularity Comparison
(All values based on net D - 1.0.)
Film Combination 4400
+
8430
4400
+
8430
No. of 8430 Duplicates 1
2
3
1
2
3
Resulting UD (calc.) 0.057
0.061
0.065
0.031
0.031
0.043
Film Type
O'D
4400
0.052
4404
0.023
8430
0.021
Table 3-5. Resolving Power Comparison (c/mm)
Test
Target Contrast
LIBR Dupe-Positive
360
310
150
1st gen. Dupe (Cony.)
360
360
180
2nd gen. Dupe (Cony.)
220
190
120
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
several disadvantages which makes it undesirable for many applications. In order
to obtain clear highlights, a combination of heavier than normal exposure and
high silver halide solvent developers are employed. The increase in exposure
generally produces a spreading of the image which degrades image quality. The
addition of the silver halide solvent reduces the density range. Such reversal
processes also reduce the control which can be exercised over the sensitometric
characteristics of the duplicate. The requirement for development to completion
in the first developer reduces the ability to significantly change the sensito-
metric characteristics in the second developer. For aerial photography, this is
a serious disadvantage since changes in atmospheric attenuation can drastically
alter the contrast range recorded on the original negative.
Utilization of the Itek LIBR (Latent Image Bleach Reversal) processes would
appear to eliminate many of the disadvantages associated with the above mentioned
techniques. Reversals obtained with LIBR have the following basic steps: image-
wise exposure, bleach, clear, uniform re-exposure, develop, fix, wash, and dry.
It is important to emphasize that with LIBR the first developer is eliminated,
and the first chemical step is the bleach. Since there is only a single develop-
ment, which produces the positive image, LIBR reversals have the ability to pro-
vide sensitometric control comparable to that obtainable with a normal negative-
positive processing technique (i.e., develop, fix, wash, and dry). From a sensi-
tometric standpoint, then, the LIBR process combines the best of both approaches
to photographic duplication. The sensitometric control of a single development
process, and the single printing stage of a direct reversal process.
From the standpoint of image quality, LIBR also appears to offer considerable
advantages. A summary of the image quality tests to date is given below.
Resolving power tests were conducted on 8430, LIBR, and the conventional
duplicating process (negative-positive). For all tests, identical precision
printing and exposing conditions were used. The results reported in Table 3-5
are the average of 18 replicate tests at the optimum exposure conditions.
From the tests in Table 3-5, it is obvious that the LIBR duplicate maintains
significantly better resolving power than the second generation duplicate. At
high contrast 8430-LIBR maintains 140 c/mm greater resolution, at 6,3:1 - 120
c/mm and at 2:1 - 20 c/mm.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
The resolving power results require some explanation. The 360 c/mm reported
for both the 100:1 and 6.3:1 contrast tests for the first generation duplicate most
likely indicates that the limit of the printing system was 360 c/mm. It is ex-
pected that with a better printing technique, the high contrast values for both
the LIBR and first generation duplicates would be higher. Secondly, it is real-
ized that the second generation conventional values are probablylower than what
could be obtained with a better printing technique. However, the important com-
parison is between LIBR and first generation duplicate, and LIBR maintains nearly
as good a resolving power as the first generation duplicate.
A RMS granularity comparison was made between LIBR and the conventional proc-
ess. Three conditions were tested with 8430, which are as follows:
1. The standard LIBR process.
2. The first generation conventional duplicate processed 6 minutes in D-19.
3. The second generation conventional duplicate processed 6 minutes in
D-19. These samples were prepared by printing the first generation uniformity
samples onto 8430.
All uniformity samples were carefully prepared to insure that no random un-
iformities, dust, or dirt were present. The results are shown in Fig. 3-11.
This figure shows that the LIBR processed sample produced the lowest granularity
of all the conditions. Particularly significant is the comparison between LIBR
and the second generation conventional duplicate, where LIBR shows a 75 percent
lower granularity. The increase in the granularity of the second generation du-
plicate is to be expected on the basis of the statistical nature of granularity.
As previously discussed, when combining two standard deviations, even when
equal, the resultant standard deviation will always be higher.
The MTF test was conducted, as with the other parameters, as a comparison
between LIBR and the conventional two-stage duplicating process with type 8430
film. The test was conducted by printing a photographic knife-edge, and evalu-
ating the resultant edge by graphical techniques (see Scott, Scott, Shack, P.S.
and E. 7:345-249, 1963).
Fig. 3-12 is a plot of the MTF data for the several tests. The results show
that the MTF of the LIBR process is slightly lower than that of the conventional
3-54 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
%W W
0.025
SPECIAL HANDLING
Net Density
Fig. 3-11 - RMS granularity comparison of LIBR versus conventional
processing of 8430 film.
^ Second generation conventional printed with the granularity
samples from the first generation (8 min D-19)
? First generation conventional (8 min D-19)
o LIBR process
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
? 0.4
SPECIAL HANDLING
^ Ori
a Fir
+a LIB
o Sec
ginal phot
9t generat
process
nd genet
ographic
ion conve
tion cony
Bdge
tional
ntional
40 60 80 100
Spatial Frequency, cycles per millimeter
Fig. 3-12 - Modulation transfer function comparison of LIBR versus
conventional processing (developed for 6 minutes at 68?F in D-19)
of 8430 film; MTF curves have had the transfer function of the micro-
densitometer removed.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
%W %0
SPECIAL HANDLING
first generation duplicate. The results presented here are purely relative since
we could do no better than the quality of the original photographic edge. The
first generation and the LIBR duplicate are indicative of the relative loss in
modulation due to both the printing technique and the scattering in the emulsion.
The second generation duplicate is indicative of the loss accrued by the second
printing operation, plus the scattering of light in the first generation print
during the printing operation. The loss from the first generation to the second
generation is greater than would be intuitively expected. This loss can be ex-
plained by several reasons.
1. The loss is, in fact, this great, and the multiple scattering that takes
place during printing is a significant factor. During the first printing opera-
tion, the emulsion is being exposed by essentially parallel light. However, dur-
ing the second printing stage the emulsion is exposed by partially incoherent
light, due to the additional scattering that takes place in the degraded edge
used to make this second dupe.
2. The sample size used here may be too limited, and the second generation
curve presented may not be representative of the average loss that would occur
over several samples.
3. The theory of transfer function is not adequate to handle such a second
stage duplicating process. These tests were made by exposing edges, and not in
the more conventional manner of exposing sine waves. The theory of edge trans-
formation to obtain the MTF is not as well developed as is the sine wave tech-
nique. This could mean that there is inherent error in the results that is not
well understood at this time.
Regardless of the difficulties involved in obtaining the MTF from an edge
trace, it is felt that the data is, on a relative basis, correct. The important
conclusion is that the LIBR duplicate produces nearly the same MTF as the first
generation conventional. This is, of course, as would be expected since all test
conditions, except chemical processing, were identical for these two samples.
Even assuming that the results for the second generation duplicates are too low,
they could only approach the MTF of the LIBR process by being printed in a system
with no transfer function loss (i.e., straight line at 1.0 modulation). Such a
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
printing system is impossible to obtain. Therefore, it is concluded that the
LIBR - MTF is sufficiently good to always be better than a second generation
duplicate.
3-58 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
%W %W
SPECIAL HANDLING
In extracting data from photography, the most important instrumentation
to be considered is a viewing configuration. Viewers can generally be classi-
fied by their function as screening viewers, detail viewers and special viewers;
these three functions can be accomplished either by an optical or by an electro-
optical means. Table 4-1 presents the basic characteristics of these three
types of viewers.
4.1.1 Screening Viewers
Screening viewers can rapidly view photography to make gross determinations
of the usefulness of the imagery, that is, to separate out imagery affected by
clouds, water, or any of the many possible cases of abortive photography.
A basic screening viewer is the common light table. The most refined
screening viewer is a continuously variable magnifying rear-.projection viewer;
most of the commercial available viewers in this class are generally too compli-
cated with respect to film loading, and too large to be of any practical oper-
ational use.
The screening viewer should be easy to operate, that is, it should be
simple to load, have human factored viewing (image position and illumination),
have a large range of transportation speeds, and not damage the image in any way.
The range of viewing magnification should be a function of the resolution
characteristics of the photography. As an example, photography that has 20 lines
per millimeter resolution should not be viewed at 20X. Most operational photog-
raphy is about 50 lines per millimeter, and the best possible operational
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Table 4-1. Photographic Data Reduction Viewers
OPTICAL ELECTRO-OPTICAL
SCREENING
Magnification
2X - 3X
2X - 8X
2X - 8X
Size
3'W x 2'H x 2'D
3'W x 6'H x 8'D
4'W x 4'H x 3'D
Volume
12 cu.ft.
144 cu.ft.
48 cu.ft.
Weight
100 lbs.
1000 lbs.
700 lbs.
DETAIL
Magnification
2X - 20OX
2X - 200X
2X - 20OX
Size
3'W x 2'H x 2'D
3'W x 6'H x 8'D
4'W x 4'H x 3'D
Volume
12 cu.ft.
144 cu.ft.
48 cu.ft.
Weight
100 lbs.
1000 lbs.
700 lbs.
SPECIAL
Magnification
Size
6'W x 6'H x 8'D
6'W x 4'H x 3'D
Volume
288 cu.ft.
72 cu.ft.
Weight
1500 lbs.
1000 lbs.
4-2 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
obtainable photography may have 200 lines per millimeter. The resolution of the
human eye is approximately 10 lines per millimeter. Therefore, the limit of
viewing magnification is from 5X (50/10) to 20X (200/10), but this is the ex-
treme limit. The practical screening viewing range should be some factor of
this limit. A subjective. estimate for this factor is that it would be a little
less than 1/2, or a magnification range of 2X to 8X.
A screening viewer should also have a film marking and/or and editing
capability. Film marking is useful in recording some of the gross decisions
made by the screening operator such as the marking of those frames that should
be removed from the roll, or that are not to be reproduced because they may con-
tain only clouds, open water or useless information. Film editing which is
similar to marking can be used for correcting any textual errors in the data
block, frame count (if some frames were removed), and for other types of
pertinent data that should be recorded with the screened photography.
A coarse measuring system is not necessary, but may be helpful in screening
certain specially collected photography.
All of the above capabilities may be accomplished either by an optical or an
electro-optical viewer. However, in a screening viewer a very useful function
is to be able to view a negative as a positive and have a readout of the photo-
graphic gamma required to photographically produce a similar positive image.
This function can only be achieved with an electro-optical viewer. Other than
this capability, and a possible reduction in size, an optical viewer is probably
best for the screening function.
4.1.2 Detail Viewers
Detail viewers are used by the photointerpreter to aid in the extraction of
information from various sensor materials.
Detail viewers are generally large magnification viewers. They can be
either non-stereo or stereo, rear projection or direct viewing, monocular or
binocular, optical or electro-optical. A large range of magnifications are
usually available and can be accomplished in either step magnification or zoom
magnification. The simplest detail viewer is a B & L zoom microscope with a
light table.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
The detail viewer should be capable of providing the photointerpreter with
the least resolvable element on the photograph. Therefore, the highest magni-
fication should be slightly lower than the highest resolution of the photography
(200X) so as to provide the photointerpreter with more than he can actually use.
The general requirement of most photointerpreters is to have a zoom magnification
system so that the photointerpreter may zoom in on a point of interest. This
zooming technique allows the photointerpreter to continually watch the point of
interest as it is being magnified, even though he may only use the lowest and
highest magnifications.
The detail viewer should be capable of viewing roll film, chip film and
glass plates with film transport speeds that are relatively slow, and which
would not damage the film in any way. Also, it should have an accurate mea-
suring system with some computational ability to convert photographic measure-
ments into useful ground measurements.
The data block should also be available as the photointerpreter may at
times require all available information concerning a particular sortie.
The detail viewer probably should be a combination of a low magnification
rear projection system and a high magnification direct stereo viewing system.
The low magnification rear projection offers a quasi-screening function to the
photointerpreter since he will not look in detail at all of the imagery. The
high magnification direct stereo viewing will provide the most photographic
information at the best possible resolution with the smallest amount of instru-
mentation. Stereo viewing requires some flexibility in differential magnifi-
cation between viewing channels and some image rotation to perform relative
orientation of the two stereo images.
The detail viewer should probably be an optical viewer as electro-optical
viewers tend to degrade resolution. However, the screening function of a detail
viewer can be very conveniently handled by an electro-optical viewing system.
4.1.3 Special Viewers
Special viewers are a catch-all but do satisfy a very important function in
photographic data reduction. These viewers are used for change detection, image
4-4 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
ire
v4so
SPECIAL HANDLING
correlation of other sensors, image manipulation, and permit other possible
techniques for the extraction of information from photography. Most of these
viewers would probably be electro-optical viewers, because of their, required
versatility. These types of viewer will be classified by their use and will be
individually discussed.
Change Detection Viewers
These viewers are designed to present in one mode or another those image
areas that have changed between two different sorties. A viewer of this type
was developed about 40 years ago to detect subtle changes in bacteria growth
under laboratory conditions by superimposing a "before" photograph projected in
red and an "after" photograph projected in green. Upon superimposition, no
change will appear yellow, a "before" change in red, and an "after" change in
green. This technique is still very practical for photography taken of the same
area from the same exposure station; however, an aerial collection system cannot
satisfy the same exposure station requirement. Therefore, a change detection
viewer must have some rectification capability.
The resolution of change detection or the least element of change is usually
relatively large, thereby permitting a low resolution viewing system. Electro-
optical viewers can easily satisfy the rectification requirement and are gener-
ally capable of resolving the least element of change. As an example, a change
detection viewer design could consist of two matched (sync and linearity) CRT/
PM inputs and one video tube display. A display of the subtraction of the two
video waveforms of each image being analyzed for change would indicate a mis-
registration due to tilts and changes. With the ability to rectify, an operator
could minimize the misregistration by trial and error until only the changes are
displayed. The Itek ARES viewer can easily perform this entire operation auto-
matically.
Image Correlation Viewers
This type of viewer would be the most important viewer in a multisensor
reduction system. Presently, a viewer of this type does not exist. As with
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W71 7W W - -W
SPECIAL HANDLING
the change detection viewer, the resolution requirement for this viewer is
relatively low, and can be satisfied be an electro-optical viewing system.
Because of vastly different multisensor image collection systems (IR, SLR,
photography) this viewer should be under computer control for image matching.
Once common image areas are identified and matched each record could be dis-
played in a separate color (e.g., IR in Red, Visual in Green, and SLR in Blue)
on a conventional TV color tube for interpretation. Once -a photointerpreter has
learned to interpret this type of color display, a very quick and accurate
interpretation could be made.
Image Manipulation Viewer
This should be an electro-optical viewer which will allow manipulation of
the electronic waveform with video amplifiers for extracting various types of
information.
4.2 REFERENCE MATERIAL
Photointerpretation keys are seldom used in an operational environment
because they are generally of a different area, difficult to handle, and diffi-
cult to update. In addition, the current format for photointerpretation keys
is such that only when an area is almost completely identified is the key help-
ful. In brief, present photointerpretation keys are not compiled properly and
are not easily accessible for photointerpretation.
Photointerpretation keys should be under computer control, so that a photo-
interpreter can ask a few pertinent questions, with some initialization parame-
ters, and have displayed at an appropriate scale the most probable key. This
operation would be similar in many respects to computer-aided managerial
decisions that many large corporations are currently using. A better analogy
might be the construction of a criminal's face by an artist from an eye-witness
verbal description, basically atrial and error method.
4.3 OTHER COVERAGE
The use of other coverage is a storage and retrieval, and an adequate dis-
play problem. The chip concept for the solution of this problem creates a very
difficult material handling problem, whereas the roll concept creates an awkward
updating problem.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
NW %W
SPECIAL HANDLING
The data block is very useful in the proper classification of imagery for
storage and retrieval. Generally, the data block is not too valuable in photo-
interpretation as a good photointerpreter can extract a majority of the data
block information from the imagery. However, the data block is very valuable in
some cursory mensuration techniques.
The available photographic records will be defined as those records that
lie within the range of silver halide sensitivity, i.e., 3400 A to 9000 A. This
range can be subdivided into three operable regions - 3400 A to 4200 A (ultra-
violet); 4200 A to 7000 A (visible); and 7000 A to 9000 A (infrared).
4.5.1 Ultraviolet Region (3400 A. to 4200 A.)
The collection of electromagnetic energy reflectance in this region is
altitude and optically limited, that is, energy in this region can only be
imaged by certain lenses, and can only be recorded at low altitudes because of
atmospheric attenuation. However, the variation in reflectance of various ob-
jects in this energy region is rather spectacular. As an example, vegetation
is a complete absorber whereas most hard surface objects are almost always a
complete reflector of ultraviolet. A low altitude photograph in this energy
spectrum looks very much like a planimetric map since all vegetation is un-
recorded.
4.5.2 Visible Region (4200 A to 7000 A)
This is the region in which the human eye is most sensitive, and in which
most of the physiological studies have been conducted. This is also the region
in which color film is sensitive.
Many studies have been conducted in attempting to determine the value
between performing interpretation with color photography and black and white
photography. In general, the results of most of these studies indicated that
color photography was considerably easier to interpret with some improved
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
detectability; however, with color film there is a considerable loss in image
resolution and considerable difficulty encountered in processing the three-
emulsion layer film and handling the processed film since the three-emulsion
layer film tends to curl with age. As a result, the use of the additive color
technique is being emphasized.
The additive color technique was originally demonstrated in 1866 by
Thomas Young who used a three-color process. With the development of motion
pictures, considerable effort was placed on a two-color additive color process.
The optimum two-color process consisted of a 4200 A to 5600 A (cyan) filtered
black and white image and 5600 A to 7000 A (orange) filter black and white image.
The superimposition of these two records give excellent color with a minimum of
effort. This combination is also suited for aerial photography since it collects
less of the blue light (3900 A to 4400 A) which tends to be attenuated by the
atmosphere.
With this two-color process, a color image is obtained by the superimposi-
tion of two appropriately filtered high resolution black and white images,
thereby providing a truly optimum color photointerpretation system (high reso-
lution color).
Every surface has a completely unique electromagnetic reflective charac-
teristic for wavelengths from a few meters in length to wavelengths a few hun-
dred angstroms in length. Some investigators have been promulgating the thesis
that several narrow passband filtered photographs should be collected to further
exploit the reflective characteristics of objects in the visible spectrum.
However, the portion of the electromagnetic spectrum in which photographic tech-
niques and the human eye is sensitive is very small with respect to the total.
It would appear that it would be more practical to collect electromagnetic
reflective characteristics over a larger span of the spectrum and reduce this
collected data to the visible portion of the spectrum for analysis since the
human eye is faster and far more accurate than any other device in sensing slight
changes in the visible portion of the spectrum.
4-8 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
VW
N%W
SPECIAL HANDLIN
G
The flow of data reduction is fairly well defined up to the point of inter-
pretation which is as follows:
1. Collection - Collection of all electromagnetic reflection from an area
of interest in one form or another by various sensors, except for ELINT which is
an electromagnetic radiation sensor.
2. Real Time Display - All electromagnetic reflection sensor systems can
be displayed in one visual form or another. Two of the more feasible means of
display are the CRT with a moving phosphor belt and the more common line-scan
CRT with a rapid processing photographic viewer. The UV, IR, SLR, and visible
portion of the spectrum are easily handled with each of the above systems;
however, the photographic imaging system requires rapid inflight processing
prior to viewing in real time.
The photographic collection system is unique in its geometric quality and
in its resolving capabilities with respect to the other sensors. It is question-
able whether this high quality sensor should be subject to a generally erratic,
rapid processing system so that it may be quickly viewed in a real time mode
with other sensors that are distorted and of low resolution. With this reason-
ing, it is suggested that real time analysis be confined to SLR, IR, the visible
spectrum and W which incidentally cover the electromagnetic full-spectrum. The
visible spectrum could be spectrally filtered and displayed on a color TV tube
for real time color analysis.
3. Recording (film processing for photography; signal processing for other
sensors, and recording on film and film processing) - This stage of the data
reduction is the most critical and, unfortunately, in many cases the weakest link.
Little is known on how to process film to obtain maximum information (which may
not be maximum resolution or contrast, etc.). This weak link also exists, and
is possibly more severe in the photographic recording of IR, UV, visible spectrum,
and SLR, as the entire success of a collection dependent upon the adjustment of
the line scan CRT and the V/h setting on the film drive. With these sensors it
would be advisable to record them in their original recorded form, that being an
electrical waveform. Then, the electrical waveform could be monitored as it is
being recorded on film to assure good reproduction.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
4. Photographic Data Analysis - This stage contains all of the viewing re-
production, storage and retrieval, and other data handling equipments, and where
proper equipment design and employment are of utmost importance.
4-10 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
%W %W
SPECIAL HANDLING
Tactical reconnaissance photo materials are recovered in a variety of
environments depending on the particular tactical situation. The recovery
location may have located on it a full established fixed facility, a mobile
facility which provides a semicontrolled environment in which equipment similar
to a fixed facility is installed, a building which has been expediently con-
verted and has installed in it much of equipment which is normally found in a
fixed facility, or a field operation set up in tents or other provisional
shelters. The degree of control over product quality and the rate of produc-
tion will be governed by the environment and the installed equipment.
Film recovered from a photo reconnaissance mission must be processed.
Even inflight-processed material must be washed and dried if it is to be re-
tained. In a substantial operation where large amounts of film are received,
continuous processing is required to meet the established schedules. Produc-
tivity and speed depend on the rate at which the processor operates.
Figure 5-1 allows calculation of productivity based on film length for some
representative speeds of machines in use today.
In some field operations it may not be possible to install continuous pro-
cessors. Film then must be processed roll by roll in tank-rewind processors.
These devices have two capacities - 300 feet and 600 feet. Rolls are processed
one at a time, and a total time of 40 minutes is required for processing and
washing. A separate dryer is required to provide a dry original negative; this
operation entails 20 to 30 minutes.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Fig. 5-1 - Film processing production chart.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
Current aerial camera equipment accepts film rolls 500 to 6000 feet in
length with intermediate increments of 1000 and 2000 feet. These cameras are
usually installed and operated in multiples. In this case, and where many
reconnaissance aircraft are recovered in a short period, the time to complete
processing could be substantial unless multiple processing stations are avail-
able.
Any roll of film identified as having the most critical information is pro-
cessed first. It may be important to determine the amount of time required to
get the first frames of photography from this roll. If the film is processed in
rewind tanks, the entire process from insertion through to the first frames
coming off the dryer make the elapsed time to first product about 45 minutes.
If continuous processors are used, total time to the first product will depend
on total path length and film transport speed. Figure 5-2 presents a rapid
means for determining these data. The range of path lengths and transport
speeds used in the figure are representative of continuous processors available
today.
Applications of special processing technique, e.g., Bimat, are finding in-
creased use in tactical reconnaissance. A recent Air Force contract resulted in
a Bimat-equipped van which enables processing and duplication of five-inch film
as the van travels from the pickup point to the photointerpreters area. The
results produced are adequate for most tactical reconnaissance photography, al-
though not of the highest quality.
Preparation of material for use by interpreters, strike crews, for briefing
purposes, and for reference requires a variety of photo duplication equipment
which can perform enlargement, reduction or same size prints from original nega-
tive materials. These duplicates can be either opaque or transparent positives,
duplicate negatives and, depending on their use, duplicated on step-and-repeat
or continuous duplicators.
SPECIAL HANDLING 5-3
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Fig. 5-2 - Time to first access.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w NW
SPECIAL HANDLING
5.2.1 Continuous Printing
The requirement to duplicate an entire roll of material calls for the use
of continuous printers. A representative list with their operating character-
istics is shown in Table 5-1. The type of printer used depends on the task to be
performed, the environment and the quality of the output. Printing rates as well
as quality vary with each device. Some devices, e.g., C-1B, are configured for
field use, while others are only for fixed facility.
5.2.2 Step-and-Repeat Printing
This type of printer is employed when only selected areas of the photography
are to be reproduced, either singly or in multiple copies. The most simple of
these devices is the hand-operated contact or projection printer. Exposure
times on such devices vary from approximately 1 to 30 seconds, which results in
a manual production rate from 2 to about 20 prints per minute. A recently
developed group of printers, using roll paper or film, has a more intense light
source to permit very short exposure times and raise production to approximately
30 prints per minute.
5.2.3 Dodging
Dodging in various forms has generally been incorporated in most military
duplicators. The older contact printers were equipped with argon glow lamps
which could be extinguished individually to control the amount of light used to
form the print image. In some printers an aperture was provided to insert
translucent tissue at appropriate spots to selectively reduce light intensity.
The advent of electronic dodging has provided an accurate control of light in
printers. Military printers, such as the CP-10 (contact step-and-repeat),
B-10A (projection step-and-repeat) and the SP-1070 (1:1 size continuous) in-
cluded this capability. These duplicators are generally slower than their non-
dodging counterparts with a resulting reduction in productivity.
5.2.4 Duplicate Processing
When duplicates are produced on a continuous roll of film or paper, pro-
cessing can be accomplished in the same types of continuous processors described
SPECIAL HANDLING 5-5
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Fa
C
-ra
P+
w
O
co
cq
=1
as
ti
N
ra
M
N
x
x
x
x
x
N
in
M
00
x
x
x
x
x
o
00
co
w
~
~
ti
N
sr
UJ
V)
N
,J3
fA
W
e-t
ra
ra
r-4
e-t
00
o
0
o
m
a
o
3
dam'
cep
co
o
c
d.*
C >
0
cn
vs Ui 3 V)
3 r-a o. > C. > m 3 m a 3 > a 3
0 L U ut UO U U
a+ O U U
co to ? to 0 us 0 in O
o o acv r4 00 r40O r-JCD O rI00 r100
U ra co ra ra W ra -a Co U") ra co .-a 1-4 cc - to 00
LW 'C 'L7 ^C
O N L U) Ai U) . ~.J 0 R1 co (d
U C!J B` 'r 00 41 41 cq 41 N
s= 0 00 < 0 b0 \ 0 0O \ ra U JJ O b U? 0
is
C+) t8 O C UD O C rd O G 4 ra t~f 'C ?r1 ra c R1 ?,-4 ,-4
?r-c Sri N to m ti cD Q 1-4 t- cc ?ra r--4e 0 ? p4 05 14 .0 i 3 UU cn ,In N u
N w
a)
U 44 C C C
cd ?,4 ?r4
U! CO Q F! W W
U tY, O 0 to Cam) C') 44)
to ts. 00 C. M ?.
bO
J.J
cd .r.,
0
~ 0 l LO
p4 r-a -4 C' ra M
(0 Q
94
ra bO
VJ 00 C C
u O V o 0 O 0
ra C) C Li O Ai a JJ ^~ JJ O JJ JJ
cJ. C U 'C3 U C U I U C U u
co T ra td ?r+ td ?rc to ?.4 cd ?rI td ,a td
H H JJ tJ O JJ L 4.J J.J L L 4.1 LJ JJ iJ
C C" C C C C C C G C C C
O O O O O O O O O O O O O
U U rd a t) U U U V U u u U
N
4)
u
60-A 0
r\-J iOJ O C d L
CU U 4 0 'C C td C
0 44 t L O U) 5 .Y. Ora x to 5 `. iJ
O u u C JJ tt! U A cd bO +J (d ?r4
ra 4) C O 0 'L7 C ca U) 'C id UJ 'U CA r4
Ra ra O U cd O O 41 cd O ?r4 CO O -d ?ri
to W UvW at. 0- C W Z W .W u z
5-6 SPECIAL HANDLING
C
C
N
1.4
?P4
?ra
W Q
4-1
44
JJ *C
LO 4))
LO
4))
00
44 E
,i a
ra
C.I.
H
cd
1J
-,4
cC -I
Z
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
previously. Because of the sensitometric characteristics of the duplicating ma-
terials, they can be processed at about twice the rate of original film. The
dried roll of duplicates must, after processing, be cut into individual segments.
A variety of automatic cutters are available which can cut these rolls at rates
up to 50 feet per minute.
The use of sheet paper or film requires tray processing. This procedure is
tedious, time consuming and messy. A line of printer, processor and fixer usu-
ally requires three people with one processing line accommodating no more than
two printers. After fixing, a lengthy wash is required; drying is accomplished
on a drum dryer in a dehydrator cabinet. Several machines have been developed
recently to overcome these problems. Eastman Kodak, Pako and Itek have developed
devices which process these materials automatically enabling the entire oper-
ation from dry to dry to be accomplished by one man. These machines generally
operate at from 2 to 10 feet per minute.
5.3 PRODUCTION VOLUME
Reconnaissance aircraft are currently equipped with a variety of cameras,
each having its own characteristics, and each having different film capacities.
Film capacities vary in width from 70mm to 9 1/2 inches, and in length from
100 feet to 12,000 feet, the latter in two reels of 6000 feet each. Cameras
may be installed and operated in multiples or individually depending on the as-
signed mission. During periods of extreme activity, it is not uncommon for the
total complement of aircraft to fly at least one mission each, using all cameras
on each mission. In this circumstance, when film is not processed inflight, it
is entirely possible for a tactical reconnaissance operation to produce from
20,000 to 40,000 feet of film per day. When this material must be processed and
printed, as well as analyzed within a two-hour period (the accepted time limit
for completing the processing and analysis of tactical reconnaissance material),
it is evident that a large complement of machines and personnel must be avail-
able, and that the process must be accellerated by a combination of high speed
processing/printing devices and automation or a combination of both.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
In addition to processed and titled original negatives, the tactical photo
operation may produce a variety of products to support battle activities.
5.4.1 Photo Interpretation
In many photo activities, where time for reduction of photography to intel-
ligence is extremely critical, interpretation will be accomplished from the orig-
inal negative. Interpreters must be trained in this form of interpretation, but
once proficiency is achieved, the results are nearly equivalent to those obtained
working from positives.
Tactical interpreters, in general, have worked from positive prints. Where
the reconnaissance system produces material of low resolution, paper prints are
adequate to retain all available detail. Producing paper prints with continuous
or step-and-repeat printers is rapid and requires only nominal control. With
the advent of high acuity systems, a large percentage of the available detail in
the negative can be lost printing on paper. As a result, tactical operations
have generally converted to continuous printing on high resolution positive
transparent materials. The resultant transparent positive is viewed by medium-
magnification rear projection viewers, or directly with high magnification
microscopes. In a limited number of cases, instead of continuous contact print-
ing, the transparencies are either magnified or reduced to convert all materials
to a standard handling size. This activity could result in the requirement for
one or more duplicates requiring not only the printers, but the processing de-
vices required to develop, fix, wash and dry the material. The machine require-
ment is directly dependent on the volume of materials received, the numbers and
types of duplicates required, and the time available to produce the required
products.
5.4.2 Reference Material
Duplicate transparencies or opaque prints are produced to be used as com-
parative reference with newly acquired coverage. In addition, copies of anno-
tated photography and maps are produced as evaluated reference. Sizes of these
products will vary with the particular reference system established.
5-8 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
Transparencies may vary in size from the recently established 70 by 100mm Unit
Record (film chip) to a full frame of original photography. These records are
indexed and stored in dossiers or in file drawers to be extracted for comparison
when new cover is received. Often prints of complete or partial frames are made
for storage in target dossiers as part of the total complement of target refer-
ence material. The index used can be the BE (Bombing Encyclopedia) Number, the
Tactical Target Index Number, or the Target Data Inventory (TDI) Number.
5.4.3 Mission Folder Material
As a part of the mission folder organized to provide reference for the com-
bat mission, photo reproductions of annotated maps and photos are produced.
Response speed for reproduction of these materials must be minimized to support
the "quick reaction" capability of combat aircraft.
5.4.4 Briefing Material
In addition to the intelligence extracted from aerial photographs, these
photographs are used in briefings as additional aids. Often selected photo-
graphs are enlarged to considerable size to serve as detailed illustrations for
the briefer. Prints of smaller dimensions are used in briefing packages for th
senior command staff. Magnification of lOX are not uncommon in this application
resulting in prints that are 40 inches square. This requirement imposes special
problems in terms of equipment to print and process as well as the paper re-
quired for this product.
SPECIAL HANDLING 5-9
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67BOO657ROO0300210001-4
SPECIAL HANDLING
6. HIGH RESOLUTION SIDE LOOKING RADAR
This section discusses the role of the radar sensor in an integrated multi-
sensor system for tactical reconnaissance. A side looking radar can be used to
obtain radar "pictures" of terrain features and cultural objects. In addition,
it can automatically detect moving objects and objects of large cross-section
and use this detection information to key the radar imagery of the information
obtained from the other sensors. By repetitive cover over the same area of a
tactical battlefield, changes in the location of enemy offensive and defensive
equipment and construction can be noted by comparison of the latest imagery with
that previously obtained. The proper choice of frequency can allow the penetra-
tion of foliage and camouflage. Unlike photography and infrared sensors, radar
is essentially an all-weather sensor.
With conventional radar systems, resolution is achieved by radiating a beam
sufficiently narrow that the width of the beam at the desired range gives the re-
quired azimuth resolution. Similarly, a conventional radar's range resolution
is achieved by radiating a pulse whose width corresponds to the desired range
resolution. However, by employing the relatively new techniques of synthetic
array radar signal processing and pulse compression, very fine azimuth resolu-
tion, practically independent of range, is obtained along with fine range reso-
lution compatible with high average power transmission; this discussion
considers the optimum design of a radar sensor incorporated in a multisensor
system for airborne tactical reconnaissance for the 1970 era. This is specific-
ally interpreted as meaning that the flight tests of the production prototype
will be completed in 1970. A reasonable extrapolation of the state-of-the-art
is made based upon current technology.
SPECIAL HANDLING 6-1
Approved For Release 2009/07/08: CIA-RDP67BOO657ROO0300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Electronic or coherent optical signal processing techniques may be employed
to process the synthetic array radar raw data to obtain imagery maps or to proc-
ess the data for moving target indication. With the present state-of-the-art,
electronic processing has proved less efficient than coherent optical processing
in terms of the weight and power required for processing at a specified data
rate (in resolution elements per second). Further developments in storage media
and electron beam technology may make electronic processing attractive in the
1970 time era. Even with the present state-of-the-art, electronic processing
can be used for moving target indication while in flight, although the number
of range resolution elements is restricted.
There appear to be very good reasons to perform inflight signal processing
to obtain moving target indications since this indicator can be used as an alert-
ing and keying device for the radar imagery and other sensors. However, there
appears to be little reason for doing inflight image processing just for the pur-
pose of presenting the output to the aircraft observer. This is principally be-
cause the observer does not have available the amount of time required to absorb
the information at the sensor output data rates from even one of the imagery
sensors. Therefore, the principal reason for doing inflight processing of
imagery is to shorten the time involved in transferring the imagery from the
aircraft to the carrier and then to the reconnaissance analyst. It is not clear
whether this time saving is commensurate with the additional amount of airborne
equipment required, with the resulting problems in weight, power, complexity,
reliability, and cost.
In analyzing the data link operation we must consider the types of informa-
tion which is valuable to transmit to the priority analyst in real-time. Obvi-
ously the MTI keying information is of this type. Also, the other keying infor-
mation obtained from the other sensors is of this type.
When operating at high altitude it is possible to transmit the radar signal
raw data directly to the carrier where it can be recorded and processed to form
an image. Whenoperating at low altitudes, however, it is not possible to ob-
tain direct transmission unless a relay aircraft is used. Without the use of
relay aircraft, it is necessary to record the raw data on film, develop the film,
and then scan the film data with a flying spot scanner and read out over the
6-2 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
data link when the aircraft is at the proper position. This requires a relative-
ly wide band link unless some selection is done of the data which is to be
transmitted back.
The requirements of low altitude and high altitude missions place different
requirements upon the performance of a synthetic array reconnaissance radar. For
low altitude missions the coverage is usually restricted by the shadowing effect
at low grazing angles, while for high altitude missions the coverage is usually
restricted by the amount of recording capacity that can be included in the radar
system. The maximum range for high altitude missions can be set by the amount
of average power, the grazing angle, or ambiguity considerations.
For low altitude, short range missions which map preselected targets, it is
possible to employ a conventional radar and still obtain reasonable azimuth res-
olution. In the high altitude, wide coverage mission, however, reasonable azi-
muth resolution can only be obtained by synthetic array radar techniques. For
the limited amount of coverage obtained on low altitude missions normal photoin-
terpretation techniques can be employed for target recognition. In the wide cov-
erage, high altitude missions, the amount of data per mission could saturate the
reconnaissance analyst unless some data selection is performed. This data selec-
tion can best be done by change detection techniques which call attention to
those areas where there are changes in radar return.
Because of weather attenuation, the useful radar frequency range is K-band
and below. Even at Ka-band there is fairly severe weather attenuation and back
scattering when light rain is present. At X-band almost all of the atmospheric
attenuation, weather attenuation, and clutter effects have disappeared. However,
the K-band frequency range has better contrast between such low reflectance tar-
gets as concrete, dirt, grass, etc. For low altitude, short range missions, the
effect of weather can be ignored and this would indicate that K-band would be
the best operating frequency to obtain good radar imagery. However, for long
range high altitude missions X-band is to be preferred.
To penetrate dense foliage and camouflage requires the use of low radar fre-
quencies, for example, 1-meter RF wavelength.
The time scales of importance in radar sensor design are fifteen minutes
and two hours. The fifteen minute figure pertains to obtaining information in
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
time to-change the mission of the next reconnaissance sortie or to change the
armament and mission of the attack force on the next flight. This timeliness
is especially important for fleeting targets in a battlefield environment. Be-
cause of the flight time and data recovery times involved, the fifteen minute
figure can only be met by the use of a data link to send priority information.
Because of data link bandwidth considerations, only selected data can be trans-
mitted, and because of the interpretation times involved, only data which is
not subject to long interpretation will be valuable.
For the radar sensor, the moving target indication information falls in
this category of priority information. Therefore, a real-time inflight proc-
essor for moving target indication should be employed and the output of this
processor should be both presented to the observer on the aircraft and sent by
data link to the carrier. On the carrier the priority analyst will be presen-
ted with this moving target information along with other alerting and keying
information from other sensors in order to make a decision for the next strike
and subsequent strikes.
Another form of operation where near real-time processing will be valuable
is for battlefield surveillance. To best utilize the radar in a timely manner,
it is necessary to fly at high altitudes and maintain continual data link con-
tact with the carrier. By the use of repetitive surveillance missions the
imagery data which is sent via data link to the carrier can be processed and
compared to previous imagery by change detection methods. To make maximum use
of this information for programming attack missions, the delay in this process-
ing and change comparison must be kept to times on the order of fifteen minutes
for fleeting targets.
The bulk of imagery data, especially on wide coverage missions, will be
physically recovered on the carrier. Film processors and signal processors
that obtain the output imagery used for data and target interpretation should
operate so that output images are available in under an hour, which means that
most of the reconnaissance information can be obtained in under two hours. In
cases where high quality photography has been obtained of the same areas being
covered, the radar will not be used for detailed interpretation but for keying
large cross-section targets and indicating changes in the radar imagery. This
keying data will enable a photointerpreter to inspect in detail these locations
on the photography.
6-4 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Synthetic array radar is one of the most advantageous technique for air-
borne reconnaissance. In that it is a new technique, it is described in detail
in this section. Functional elements of a radar system are shown in Fig. 6-1.
Conventional radar systems achieve azimuth resolution by radiating a beam
sufficiently narrow that the width of the beam at the desired range gives the
required resolution. Similarly, in conventional radars, range resolution is
achieved by radiating a pulse whose width corresponds to the desired range res-
olution. The result of these conventional techniques is a radar whose range
resolution is essentially independent of range but whose azimuth resolution de-
teriorates with range, since a constant angular resolution is achieved.
In many instances, a short pulse technique is adequate for the achievement
of useful range resolutions. However, in many cases the linear azimuth resolu-
tion desired cannot be achieved with particular parameter values (antenna size).
Improvement of the azimuth resolution by conventional techniques is accom-
plished by narrowing the radiated beam. Since the radiated beam is proportional
to wavelength and inversely proportional to the length of the antenna aperture,
much effort has gone into systems using shorter wavelengths and longer antennas.
Obviously, for airborne mapping applications, a limit is reached on the length
of antenna which can be carried in an aircraft. For this reason, means to
achieve fine resolution without the necessity for long physical antennas were
sought, which resulted in the development of the synthetic antenna technique.
The synthetic antenna technique uses motion of a significantly smaller
antenna together with signal analysis operations to generate a synthetically long
antenna. The analysis of the operations involved shows that a linear azimuth
resolution can be achieved which is independent of range and of frequency, and
which depends only upon the aperture of the physical antenna used. Moreover,
contrary to the conventional radar case, finer resolution is associated with
smaller physical apertures, the theoretically achievable resolution being, in
fact, equal to approximately one-half the physical aperture.
The reasoning leading to a synthetic antenna is essentially as follows.
In a linear array the beamwidth of the array factor is inversely proportional
SPECIAL HANDLING 6-5
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Waveform
Generator
Motion
Compensation
Injection
J
Motion
Compensation
Computer
Airborne
Processor
Data Selection
Display
Telemetry
Transmitter
Telemetry
Receiver
Photographic
Film
Inertial
Platform
MTI Alerting
Signal
Power
Threshold
Alerting
Alternate Transmission
Path Utilizing Inflight
Processing
Strip
Map
Output
Optical
Signal
Processor
6-6 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
to the length of the linear array, so that one achieves narrower array factors
by using longer arrays; in a physical linear array, each of the elements of the
array exists physically, and each element transmits and receives signals
simultaneously. The signals from the radiating members of the array are com-
bined so that appropriate vector summation of their signals gives the radiated
beam pattern associated with the array.
In the synthetic antenna case, however, only a single radiating element is
used. This element is moved so that it occupies, in turn, each of the positions
of a linear array. At each of these positions, a signal is transmitted and the
radar echoes are received. The received signals are put into storage and subse-
quent operation on this data is used to combine these signals in a manner ap-
propriate to synthesize a long aperture antenna. To do this, the operations
must preserve the essential vector character of the signals so that a coherent
radar system preserving both phase and amplitude is required.
It is evident that, if the signals are in storage, some degrees of freedom
available in the synthetic antenna case which are not available in-the case of a
real physical linear antenna. Since the data is in storage, a synthetic antenna
can be generated which is focused at all ranges. In generating synthetic anten-
nas focusing becomes necessary ag the desired length of the synthetic antenna is
increased. For antenna lengths shorter than some critical value, little is gain-
ed by focusing. However, for synthetic antenna lengths longer than this critical
value, focusing is required in order to achieve the resolution corresponding to
that length. Hence, two types of synthetic antennas are considered, namely,
focused and unfocused.
An additional benefit derived is that a synthetic antenna of a given length
gives a resolution element finer by a factor of two than that achievable with a
physical antenna array of the same length. This arises because path length dif-
ferences, both in transmission and reception, are effective in generating the
synthetic antenna pattern, whereas only path length differences during reception
are effective in the physical antenna case.
Finally, antenna lengths of the order of tens of feet are practical in air-
borne systems. In the case of synthetic antennas, array lengths of thousands of
feet are possible since these antenna lengths are achieved by transporting a
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
small antenna along the synthetic array. The length of the synthetic antenna
achievable depends only upon the product of beam width of the antenna and range
to the target. Thus, it is evident that wider beams are needed to achieve long-
er synthetic antennas. This is contrary to the conventional case in which phys-
ically long antennas are required to produce narrow beams.
The notion of the synthetic antenna beam first received attention prior to
1953. Since 1953, however, much more intensive activity has been devoted to
synthetic antennas. Many systems have been built and flight-tested since that
time, and flight test results of successful systems are available. Some
flight test results are shown in Section 6.1.6.
6.1.1 A Comparison of Several Azimuth Resolution Techniques
This section will compare the resolution performance capability of the
three means for achieving azimuth resolution, namely:
1. The conventional technique: In this technique, azimuth resolution de-
pends upon the width of the radiated beam.
2. The unfocused synthetic antenna technique whereby the synthetic antenna
length is made as long as the unfocused technique permits.
3. The focused synthetic antenna technique, whereby the synthetic antenna
length is made equal to the linear width of the radiated beam at each range.
As is shown in the sections which follow, the linear transverse resolution
for the conventional case is given by:
Resolution cony = ? D R . (6.1)
For the unfocused case, the resolution is given by:
Resolution R
unf 2 ,
whereas for the focused case, the resolution is given by:
Resolutionfoc 2
(6.2)
(6.3)
In the above expressions, A is the wavelength of the radar signal transmitted, D
is the horizontal aperture of the antenna, and R is the radar range.
6-8 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Fig. 6-2 is a plot of the resolution for each of these cases as a function
of radar range. This plot is for an antenna aperture of 5 feet and a wavelength
of 0.1 foot.
Conventional Azimuth Resolution - The conventional technique for achieving
azimuth resolution has been that of radiating a narrow beam. In this case the
resolution of a target depends upon whether the target is included within the
radiated beam. While some techniques exist for resolving targets somewhat less
than a beamwidth apart, it will be assumed that the resolving capability of an
antenna is equal to the width of the radiated beam at each range.
The computation of the linear azimuth resolution for the conventional case
is well-known. The appropriate expression is obtained by noting that the width
of the radiated beam in radians is given by the ratio A/D, whereas the linear
width of the beam at range R is the product of this beamwidth by range. These
considerations lead to the result described by Eq. (6.1).
For a given wavelength and antenna aperture, the fineness of resolution de-
teriorates in direct proportion to range. For a specified antenna aperture and
range, the linear resolution can be improved by using a shorter wavelength.
Conversely, with the wavelength fixed, an increase in antenna aperture is re-
quired to improve resolution.
A consideration not evident from the above discussion but pertinent from
antenna theory is that Eq. (6.1) applies only to the far-field pattern of an
antenna. The beginning of the far-field occurs at a distance min for which
D2
Rmin T
It will be noted by substitution of Eq. (6.1.1) that the finest resolution
achievable using the conventional technique is given by
Minimum Conventional Resolution = D (6.1.2)
The Unfocused Synthetic Antenna - The simpler of the synthetic antenna
techniques is that which generates an unfocused synthetic aperture. In this
case, the coherent signals received at the synthetic array points are integrated
with no attempt made to shift the phases of the signals before integration.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
1000
Curve c (focussed case)
10 100
RANGE (N.M.)
Fig. 6-2 - Azimuth resolution for
three cases: (a) conventional, (b)
unfocused, (c) focused.
Fig. G-3 - Geometry for unfocused synthetic antenna.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
This lack of phase adjustment imposes a maximum length upon the synthetic anten-
na length which can be generated. This maximum synthetic antenna length occurs
at a given range when the round-trip distance from a radar target to the center
of the synthetic array differs by A/4 from the round-trip distance between the
radar target and the extremities of the synthetic antenna array.
The pertinent geometry is shown in Fig. 6-3. In this figure, R0 represents
the range from a radar target to the center of the array, Leff represents the
maximum synthetic antenna length for which the distance from the target to the
extremities of the synthetic antenna does not exceed R0 + T/8.
From this geometry,
L )2
(Ro + 8)2 = ( Off)+ Roe.
(6.2.1)
If this expression is solved for Leff' subject to the assumption that T/16
is small compared to Ro, the result is
Leff = 41R 0T . (6.2.2)
Because of coherence, the relative phase of signals received from a target
depends on the round-trip distance to it. Consequently, the beamwidth of the
synthetic antenna is half that of a conventional array of the same length, i.e.,
(3 = radians.
2Leff
(6.2.3)
Multiplying this beamwidth by range results in the resolution given by Eq. (6.2).
As shown in Eq. (6.2), the transverse linear resolution is independent of
the antenna aperture size. For the unfocused case, fineness of resolution is
increased by the use of shorter wavelengths, but in comparison with the conven-
tional case, the improvement in fineness of resolution varies as the square root
of T rather than directly as T.
The Focused Case - An expression for the resolution achievable in the
focused case was given by Eq. (6.3). It is significant that the azimuth resolu-
tion achievable for this case depends only upon the synthetic antenna aperture
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
and that, in contradistinction to the conventional case, finer resolution is
achievable by using small rather than large antennas.
Also, significant for this case is the fact that the achievable resolution
for a given antenna size is independent both of the range and of the wavelength
used. A graph of Eq. (6.3) is also shown in Fig. 6-2. Inasmuch as the resolu-
tion achievable in this case is a constant, this curve is a straight line par-
allel to the range axis.
In order to achieve the resolution indicated by Eq. (6.3), a synthetic
antenna length given by Eq. (6.3.1) is required; namely,
_ M
Leff D
(6.3.1)
The consideration used in arriving at Eq. (6.2.2) indicated that unless addi-
tional processing were applied to the signals, antenna lengths such as those
implied by Eq. (6.3.1) could not be achieved. The processing required is that
of an adjustment of the phases of the signals received at each point of the syn-
thetic antenna which makes these signals co-phase for a given target. If this
is done, the restrictions which limited the maximum antenna length to that given
by Eq. (6.2.2) are no longer pertinent, and the new limitation on the length of
synthetic antenna achievable becomes simply the linear width of the radiated
beam at the range of the target. Note that Eq. (6.3.1) is an expression for the
maximum length in this case.
In some cases a resolution coarser than D/2 is sufficient. Then a fraction
a of the maximum focused synthetic antenna length can be used. Eq. (6.3.1)
becomes
_ a A R
Leff D
Resolution D
foc
(6.3.2)
In those situations for which the synthetic antenna length given by Eq.
(6.3.2) is less than or equal to the synthetic antenna length for the unfocused
case as given by Eq. (6.2.2), only a limited improvement in resolution is
6-12 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
achievable for the focused case. However, if a resolution finer than that given
by Eq. (6.2) is desired, then focusing must be used, and in this sense the use
of focusing, in effect, removes the restriction on synthetic antenna length which
would otherwise apply.
6.1.2 Ambiguity
The basic mechanism employed in synthetic aperture radar is to operate on a
"phase history" that can be predicted for each target in the antenna beam. Pre-
dictability, properly accounted for in the signal processing, will permit the
obtaining of resolution much finer than conventional beamwidth limited consider-
ations would indicate. Let us consider, for the moment, the phase history that
will be obtained from a point target with a side-looking, coherent radar.
A side-looking radar is an airborne radar that has the main lobe of the
antenna beam oriented normally to the velocity vector of the vehicle. A coherent
radar is distinguished from a conventional radar by a clock employed in the radar
set that permits the measurement of the total round-trip phase distance to a tar-
get (modulus 27r). The diagram shown in Fig. 6-4 shows the airplane on the left of
the picture with the antenna beam depicted. A target, whose minimum range is ro,
can be thought of as traversing the path shown as Target Path relative to a sta-
tionary airplane. The change in a observed as the target moves along the target
path is proportional to the phase output from the coherent radar. This radar is
easily calculated by
( r0 + a) 2 = rot + x2 ,
r 2 (1 + 2a) = rot + x2
o
r 0
Reduced to a first order approximation,
x2
a = 2r
0
The approximation here assumes that the distance a is much smaller than r0 which
is clearly a valid assumption when the antenna beam width is on the order of a
few degrees and the range r0 is several miles.
SPECIAL HANDLING 6-13
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
Flight Patti
SPECIAL HANDLING
Fig. 6-4 - Side looking radar pattern from aircraft.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w W
SPECIAL HANDLING
When the distance a is converted to a round-trip phase distance, the follow-
ing result is obtained:
2 7r x2
r
0
(6.4)
where T is the radar wavelength. A final change is made by our establishing a
time origin at the instant the target is exactly broadside the aircraft; and then
the distance x is merely the product of aircraft velocity and time. Thus, the
phase history is expressed as
fi(t) 27rv2 t2
A r
0
(6.5)
If required, a frequency history could be specified as the time derivative
of Eq. (6.5), or
t) = 47r v2 t
r
0
The maximum value that w can obtain is when the target just enters or just leaves
the beam. In terms of Eq. (6.4), this maximum value is
41r xmax dx
max (t) A ro dt
47rvx
max
T r
(6.6)
A maximum value for x is obtained by our considering the nature of the antenna
characteristics. The beamwidth of the antenna is roughly the ratio of the radar
wavelength to the aperture length. Therefore, the distance xmax is
_ 0
xmax 2D
where D is the antenna length. Thus we obtain
= 2v_
max ~D'
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
or in terms of frequency measured in cycles per second,
f = V
max D
(6.7)
The above discussion has led to a basic consideration that can cause am-
biguity. It is well established that unique recovery of a sampled signal can be
accomplished only if the sampling frequency is at least twice as great as the
bandwidth of the signal being sampled. When the sampling frequency is actually
less than twice the bandwidth, spectrum folding occurs, and it becomes impossible
to distinguish between different frequency signals in the sampled spectrum. In
other words, ambiguous signals are generated by the folded portion of the spec-
trum.
In synthetic aperture radar, the data processing is unable to separate sig-
nals from the folded portion of the spectrum from the undistorted portion of the
spectrum. System pulse repetition frequency (PRF), therefore, must be set at a
sufficiently large value to permit ambiguous recovery of the spectrum from any
target illuminated by the radar antenna. It is important to observe here that
the bandwidth of the signals being sampled is not merely the ratio of the vehicle
velocity to the antenna aperture length as shown by Eq. (6.7). The bandwidth
is in reality twice this value because information is collected both before the
vehicle is broadside to the target, and after this time. When the target is in
the forward half of the antenna beam, the velocity is closing, (i.e., the air-
plane is moving toward the target). During this period of time, the frequency
shift is positive. On the other hand, when the target is in the after portion
of the beam the target is receding and the doppler shift is negative. Here is a
case where it is important to be able to distinguish between positive and nega-
tive frequencies - a task that is very difficult to accomplish. To assist in
this measurement, a frequency offset is employed. The offset is set at a value
such that the lowest negative frequency is translated to zero frequency. This
amount of offset frequency will translate the entire spectrum of received sig-
nals to positive frequencies, and there will be no difficulty encountered in
separating the different portions of the spectrum of the return signal. The
amount of this offset must be at least as great as fmax determined by Eq. (6.7).
Therefore, the actual maximum frequency to be sampled by the system PRF is in
reality 2 fmax, and the minimum PRF is
6-16 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
PRF > 4v
D
(6.8)
In case the PRF is less than the above quantity, the data processing system will
be supplied with false signals that are indistinguishable from bonafide signals.
Another problem that is related to ambiguity generation sets an upper limit
on the PRF. When the inner-pulse period is shorter than the round-trip propaga-
tion time for the required range interval, two pulses can impinge on the ground
at the same time within the region illuminated by the antenna beam. When this
operation occurs, signals from targets widely separated in range will arrive at
the radar receiver simultaneously. No really workable method of separating these
range signals has been demonstrated. Therefore, the inter pulse period must be
kept greater than a value corresponding to the round-trip distance across the
range interval being mapped. This consideration leads to
T > 2 A R
'
c
PRF < c
2 0 R
(6.9)
where c is the propagation velocity.
The inequalities Eq. (6.8) and (6.9) are now combined to show the boundaries
for the permissable range of PRF:
Dv < PRF _ < 2 L R
(6.10)
When it is possible to place the PRF within the above boundaries, no diffi-
culty will arise from ambiguity considerations. In some cases, the permissable
range of PRF might become very small, or it might not be possible to satisfy the
above expression at all. These cases occur because the required resolution is
very small, and hence the antenna aperture must be small, the velocity of the
vehicle is very large, or a very wide range coverage is necessary. If one or
more of these considerations force the boundaries on the PRF to the same value,
i.e.,
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
4v c
R '
D 26
the radar is called ambiguity limited, and there is no freedom available in the
selection of various system parameters. A most important restriction imposed by
ambiguity limited performance is on the radar antenna area. Solving the above
equation for the length of the antenna, we find that
D= 8 v d R
C
(6.11)
A relationship for R can be derived which shows that R depends on the mini-
mum slant range to the target, the radar wavelength, the vertical antenna aper-
ture and the depression angle. Consider the diagram shown in Fig. 6-5. The law
of sines relates the quantities shown in this figure by
Rmin _ R
sin 4) _ sin a
As an approximation, sin a is the ratio of the radar wavelength to the
height of the antenna. Therefore, A R can be expressed as
Rmin 7\
R h tang
where h is the vertical aperture of the radar antenna. When this expression is
substituted into Eq. (6.11), it is possible to solve for the product Dh, the area
of the radar antenna:
BvRmin ?
Dh = Amin c tan 4)
(6.12)
Equation (6.12) is an expression for the minimum antenna area. This area
has been dictated by ambiguity considerations, and if the antenna is smaller
than specified by Eq. (6.12), the received signal will contain ambiguous returns.
Note that from this equation the shape of the antenna is not dictated by the am-
biguity consideration, but rather the area is fixed. Therefore, the particular
ground coverage or longitudinal resolution is not in itself fixed by ambiguities,
6-18 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Fig. 6-5 - Calculation of R.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
but these two parameters are related in a more complex manner. When the area is
fixed and the antenna is lengthened, the height must be reduced proportionally.
This reduced height will permit the illumination of a greater range interval on
the ground. On the other hand, longitudinal resolution can be no smaller than
one half the physical length of the antenna. Increasing the range coverage of
an ambiguity limited synthetic aperture radar can -be accomplished only at the
cost of longitudinal resolution. Likewise, improved longitudinal resolution can
be obtained only at the cost of range coverage.
The above brief description of the ambiguity problems in synthetic aperture
radar is only intended to point out the physical reasons for these effects.
There has been no intent to show the system degradation if ambiguities are per-
mitted to appear in the results. This subject area has been of great interest
and extensive use has been made of the ambiguity theory developed by Woodward,
Siebert, and others. Much of this material has been published and appropriate
references to these works are shown below.
1. Woodward, P. M., Probability and Information Theory, McGraw-Hill, New
York, 1953.
2. Siebert, W. Mc., A Radar Detection Philosophy, Institute of Radio Engi-
neers, Transaction of the Profectional Group of Information Theory, Vol. IT-2,
September, 1956.
6.1.3 The Radar Equation for Synthetic Aperture Radar
The radar equation for a synthetic aperture radar is somewhat altered from
its conventional form. This modified form is derived quite easily, and the der-
ivation is shown in this section. The conventional radar equation will be devel-
oped first. When a peak transmitted power Pp is radiated toward a target at
range R through an antenna of area A, the power density at the target is
P A
P = o
R2 -X2
where A is the radar wavelength. A target of cross-section a will in turn re-
flect a portion of this power toward the receiving antenna. The power density
at the receiving antenna is:
6-20 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w VW
SPECIAL HANDLING
P _ 9
and finally, the total power intercepted by the receiving antenna is the product
of the antenna area and this power density. Generally, the receiver and trans-
mitter will use the same antenna, so that the signal power is expressed as
P A2 Q
S = p
47r R4 T2
The parameter of interest here is usually concerned with the strength of the
received signal when compared to the locally generated noise. Since the noise
power generated in the receiver is represented by the product k T B where
k = Boltzman's constant,
T = Effective receiver temperature,
B = Receiver bandwidth,
the ratio of peak received signal power to average noise power is
pp A2 U
41rR4 A2 kTB
(6.13)
At this point in the development no new ideas have been presented. Now,
several ideas from synthetic aperture radar theory shall be incorporated that
will significantly modify the appearance of the radar equation. First, it is
important to recognize that the nature of the data processing amounts to a phase
rotation of the received signal followed by an integration. Therefore, in com-
puting the signal-to-noise ratio, the power received from a given target after
signal processing must be an integration of Eq. (6.13) over a specified period
of time. Since a synthetic aperture radar is a pulsed radar, the integration
can be replaced by a summation over the required number of pulses. The signal
is coherent, and the phase of each return is corrected so that the signal ele-
ments all add in phase. Thus, the time integration of signal over n pulses
would result in a signal that is a factor n2 greater than the signal power re-
ceived on a single pulse. However, noise power will also increase after
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
integrating n samples. Noise, being a random signal, will only increase by a
factor n in power. Therefore, a net gain of a factor of n will be obtained from
the data processing, and the signal-to-noise ratio after processing will be
S pp A2 6n
N p
41r R4 T2 k T B
The problem now is to determine what the number n must be. Certainly n can
be expressed as the product of integration time and system PRF. Another modifi-
cation at this point is to replace the bandwidth term by the reciprocal of the
pulse length T. With these two changes, the radar equation can be written as
S pp PRF T A2o t i Pa A2 cr t i
N p = _
4;r R4 A2 k T 41r R4 a2 k T
where ti is the integration time, and Pa is the average transmitter power.
The value of ti is found by considering the synthetic aperture. A given
resolution 6 from a target at range R indicates an angular resolution from the
synthetic aperture of
L R
The -4 db beamwidth of the synthetic aperture is the ratio of the radar wave-
length to twice the length of the synthetic aperture, or
OL 21,
Further, the time required to fly a distance L (the integration time) is the
ratio of L to the airplane velocity. This time is
R T
2 V6
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w W
SPECIAL HANDLING
The radar equation now reduces to
S
N
P A2
a
(6.14)
8ir R3 ?\ V6 k T
One final consideration deals with the parameter a*. A terrain mapping
radar is a sensor that detects far more than widely scattered targets as is the
case in an air search radar. Specification of a target cross-section does not
want to be concerned with the nature of specific targets such as trucks, tanks,
buildings, and so forth. The matter of importance is that of being able to gen-
erate a background signal against which targets can be observed. Another impor-
tant facet here is to be able to detect the absence of this background or the
presence of a specular reflector. An example is the problem of detecting a road
or an airplane runway. Hard surfaced roads and runways are nearly specular re-
flectors. To observe this type of installation, it is necessary to be able to
detect with good clarity the grass that surrounds the specular reflector so that
these interesting target complexes will appear as the absence of background.
Therefore, our problem in specifying a target cross-section is not that of re-
quiring sensitivity to detect a certain sized target such as a vehicle but
rather to detect with a reasonable signal-to-noise ratio the backscattering from
grass or similar ground cover at the maximum radar range.
Experimental analysis of this problem has been conducted in the United
States (see R. L. Cosgriff, W. M. Peak, and R. C. Taylor, "Terrain Scattering
Properties for Sensor System Design" (Terrain Handbook II) Engineering Experi-
ment Station Bulletin, Ohio State University, Columbus, Ohio). Results of this
analysis has shown that backscattering from ground cover can be expressed as a
fraction of the resolution area projected onto the line of sight to the radar.
In other words, a can be written
a = y 62 tan 0 ,
(6.15)
where y is a number less than unity, and (P is the grazing angle. It has been
found that y for short cut grass is approximately 0.02. In practical cases for
the computation of performance at maximum radar range, the angle 4 is merely the
minimum depression angle of the radar. As a final step, the substitution of
Eq. (6.15) into Eq. (6.14) yields:
SPECIAL HANDLING 6-23
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
E- _
SPECIAL HANDLING
S Pa A2 y6 tan N
87rR3 7NvkT
(6.16)
The radar equation for synthetic aperture radar points out several of the
remarkable characteristics of this type of radar. Probably the most notable
change is the range cubed characteristic of the radar. In conventional radars,
the signal-to-noise ratio varies inversely as the fourth power of range where in
the synthetic aperture system the signal-to-noise ratio varies inversely as the
cube of the range. Therefore, there is a great savings in required transmitter
power to obtain a given level of performance.
Parameters that do not appear directly in the usual radar equation are the
resolution 6 and the velocity of the vehicle v. Here observe that the per-
formance varies inversely as the speed of the vehicle and directly as the resolu-
tion. Therefore, fine resolution performance from a radar in a high-speed vehi-
cle will require more transmitter power to obtain a given signal-to-noise ratio
than would be required to deliver the same level of performance in a slower ve-
hicle. This effect can be considered as being more than offset by the ambiguity
limitation. Recall that the minimum antenna area is proportional to velocity.
Therefore, in a strictly ambiguity limited situation the value of antenna area
dictated by Eq. (6.12) could be substituted into Eq. (6.16) to indicate the
signal-to-noise performance that includes the ambiguity consideration. This
substitution might be well justified if all radars were to be ambiguity limited.
In general, however, a radar limited in performance by ambiguity considerations
is rarely found. Separation of Eqs. (6.12) and (6.16) is important in most
cases.
6.1.4 Recorder Design Considerations
Data Storage Techniques - There are available a number of recording tech-
niques which include magnetic tape, CRT writing on film, and direct electron
beam writing on film, thermoplastic and electrostatic recording materials. Only
electron beam recorders (including indirect CRT recording) have successfully
recorded and read out information over tens of megacycles bandwidth with good
reproduction quality.
6-24 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
The direct electron beam recording technique which has received the most
attention in recent years is electron beam recording on thermoplastic material.
Electron beam recording on other materials in which local changes in the index
of refraction, polarization, thickness, etc., are induced have been studied to
only a limited degree. Electron beam techniques also have been tested in which
an electrostatic charge pattern is written on a material such as selenium, after
which readout processing similar to that used in xerography can be used. None
of these approaches of direct electron beam recording, however, has progressed
to the point where they can be proposed with confidence for system application
in a short time period. Electro-optical recording techniques employing cathode-
ray tubes, on the other hand, have been tested and proven successful. Other
methods of recording may hold more promise for the future, but are not considered
competitive at the present time.
CRT Recording on Photographic Film - A cathode ray tube-film recorder con-
sists of a high resolution CRT with its associated electronics, film spools and
a film drive with its electromechanical drive control. Inputs consist of timing
signals from the synchronizer, receiver coherent video, an aircraft along track
velocity signal and various control signals. Two different optical configura-
tions can be employed to transfer the CRT trace to a photographic film. One
uses an optical lens system with a conventional CRT face plate, while the other
eliminates the lens system by placing the photographic film in direct contact
with a set of fiber-optics which is imbedded in the face of the CRT. The prin-
ciple advantage in using fiber optics is that a greater percentage of the light
generated by the phosphor actually reaches the film. This produces increased
exposure, allowing slower film to be used to improve the resolution. A sketch
of a recorder is illustrated in Fig. 6-6.
Recorder Parameters - The recorder parameters are determined from the air-
craft along track velocity v, the required resolution 6 and the transfer charac-
teristic of the CRT - film combination chosen. This transfer characteristic re-
lates the input signal to the actual resolution on the film measured in terms of
special frequency for a given contrast level.
For a specified minimum contrast ratio, let the highest spacial frequency
that can be resolved f. Then the resolution in range on the film, in the
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Fig. 6-6 - Recorder configuration.
6-26 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
width dimension of the film, is Ar = d/f if one resolved line pair is allowed
for each pulse length. The total range interval recorded on one film is then
R=Wf d, where
P = total recorded range interval,
W = usable width of the film,
and Sr = actual range resolution (corresponding to one pulse length).
To avoid signal ambiguities, the portion of the signal doppler history
which is used for processing usually goes from zero doppler to some maximum
doppler frequency fd where
_ 2vL
fd T R
Defining ba to be the -4 db azimuth resolution, then
fd
V
S
a
6a 2L
If an offset fa from zero doppler is used then the maximum doppler frequency
that must be recorded would be fd + fa.
The speed of the film in the recorder v' is expressed in terms of the maxi-
mum doppler frequency fd and the maximum resolvable spacial frequency
, - fd v
v - f = f ba
The length I of the signal history which is to be processed is then rela-
ted to the synthetic array length L by the scale factor v'/v:
= v' L L = T R
v f a 2f6a2
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
6.1.5 Optical Data Processor
In one mechanization of fine resolution radar using the synthetic antenna
technique, a coherent optical data processor is the key component. This proc-
essor accepts as its input a signal history film which is the initial output of
the radar system and operates upon it to expose a second film which becomes the
fine resolution radar map. The signal film format is one in which signals from
increasing range occur across the width of a film while successive returns from
a given target occur along the length of the signal film. The signals are re-
corded upon -this film range-trace-by-range-trace, side-by-side, to build up the
signal history film.
The format of the output film is similar in that targets at increasing range
are also located along the width of the film. However, as a result of the proc-
essing, successive points at a given range are radar images of successive points
at that range along the radar flight path.
It is the function of the optical processor to operate upon the signal
history film in such a manner as to, in effect, generate a long synthetic an-
tenna. The result of this operation is a resolution whose fineness is that
corresponding to the length of synthetic antenna generated.
To understand the operation of the optical processor, a brief discussion of
the nature of the signal history recorded as the aircraft carrying the radar
flies past a given target is in order. This signal history resembles a linearly
frequency-modulated signal. In fact, this signal consists of a sequence of sam-
ples of the Doppler frequency generated by flying past the target under consid-
eration. Upon first entering the beam, the Doppler frequency shift has its
highest value. This Doppler frequency shift decreases to zero as the radar
reaches the broadside position, and the Doppler shift becomes negative as the
aircraft continues to fly past the target.
If the signal is recorded in this manner, one cannot distinguish positive
from negative Doppler shifts. Hence, it is usual to generate an offset frequen-
cy sufficiently high that the Doppler shift plus the carrier frequency do not
pass through zero while the target is in the radar beam. In this manner, a
linearly frequency-modulated signal which does not pass through zero is generated
6-28 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
and recorded from each target in a radar beam. The signals recorded in this
manner from a single target resemble a slice taken from a zone plate. A zone
plate has the characteristic of behaving like a lens with a multiplicity of
focal lengths, i.e., it has self-focusing properties. This property is made use
of in the design of the optical processing equipment.
Let us assume that a radar target, instead of being illuminated and re-
radiating the signal, were a coherent transmitter. Then the signals incident
at any instant of time upon the synthetic antenna would be very similar to the
signals recorded in the radar pulse-by-pulse as the radar flies by a target.
Hence, one may consider the recorded signals from a given target in question.
At first glance, it might seem that information needed to image a target is
lost in this process. Actually, such is not the case. From a knowledge of the
geometry of the radar configuration (range measurement and scale factor change)
a location can be ascertained for a small source of light which can be used to
illuminate this signal history such that the signals emerging from the signal
history thus illuminated duplicate both in amplitude and in curvature the ampli-
tudes and curvatures of the microwave signals. If one takes into account the
change in scale factor and the change in wavelength in going from radar frequen-
cies to optical frequencies, it is evident that one can place an optical system
beyond the illuminated signal history so as to create an image of the radar
target.
In essence, the optical processor behaves as indicated above. However,
several additional problems have to be considered.
The discussion above has concerned itself with creating the limits of a
single point at a single range from operations on the target signal history.
The actual signal history consists of signals from different ranges, and these
create wavefronts with different radii of curvature at the signal history film.
Consequently, the optical system must cause a wavefront to be incident upon the
target signal history whose curvature as a function of range appropriately
duplicates in analog fashion the geometry of the radar world.
There are several ways to create this curvature. One technique uses a
conical lens between a narrow slit and the signal history film.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
In addition to having targets at different ranges, the target histories at
a given range will be overlapping from targets at the same range but different
azimuths. In such cases, a detailed examination of the behavior of the system
will show that for a given segment of signal history the target histories from
the target at one azimuth will be appropriately illuminated and will cause that
spot to image whereas signals from targets at the same range but different azi-
muths will converge to points at a different azimuth. Hence, the self-focusing
properties of the target histories serve to separate targets at the same range
but with different azimuths.
Thus far, the problem of keeping the range elements separate has not been
discussed. It has been stated that the signals from a given target are found on
a line along the signal history film. In order to keep these separate lines
(range intervals) separate, the optical system following the signal history con-
sists of a combination of spherical and cylindrical lenses. The combined power
of the cylindrical and spherical lenses is such as to cause imaging in one direc-
tion in the output plane, thereby keeping the range elements separate. Thus the
optical system required to operate upon the target signal histories consists
essentially of the following sequence of elements.
1. a narrow slit in combination with a conical lens to illuminate the tar-
get history film with a coherent wavefront having the appropriate curvature as a
function of range;
2. the signal history film together with a transport to move this film
across an aperture in the optical system;
3. a combination of spherical and cylindrical optics to image the target
signal histories and to keep the range elements separate;
4. an output slit and camera for recording the fine resolution radar map.
The description of operation given above has considered the optical system
as an analog to the radar world. Alternatively, the signal processing can be
considered as one in which synthetic antennas are generated. In this case, the
sequel of signals on this target history film from a given target can be con-
sidered as being the signals which the elements of a linear array at that point
would receive. These signals have a phase shift associated with them which
points out the fact that different points along the synthetic array are at
6-30 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w VW
SPECIAL HANDLING
different ranges from the radar target. If a focused radar system were built,
the phases of these signals would have to be readjusted before vector addition.
It is the function of the conical lens to make the appropriate phase correction
to the signals before the addition. Then the signals from each point on the
synthetic array will then be added cophase.
The function of the spherical and cylindrical optics can be viewed as per-
forming a line-by-line (constant range) integration of the signals from the
synthetic array positions.
Either point of view results in the same optical configuration. A schematic
diagram of an optical processor appropriate for generating synthetic antennas is
shown in Fig. 6-7. Examples of the appearance of signal history are given in
Figs. 6-8 and 6-9.
The synthetic antenna technique was conceived prior to 1953. The earliest
flight test results using the optical processor were conducted in 1.957. Since
that time, a number of successful systems employing these principles have been
built and demonstrated.
6.1.6 Summary
The discussions have shown that the resolution capabilities of conventional
focused and unfocused antenna systems differ markedly in their characteristics.
The fineness of resolution, other parameters being fixed, varies directly as T
for the conventional case, as the square root of ?\ for the unfocused case, and
is independent of T for the focused case. Dependence of the fineness of resolu-
tion on antenna size has been shown to improve for larger apertures in the con-
ventional case, to be independent of aperture in the unfocused case, and to
improve with smaller apertures for the focused case. Dependence of the fineness
of resolution on range has been shown to deteriorate directly as range in the
conventional case, as the square root of range in the unfocused case, and to be
independent of range in the focused case. Evidently, as the requirement for
finer longitudinal resolution increases, one may be led to consider in turn the
conventional, the unfocused and the focused case, in that order.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
u G
V Ol
6-32 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
w
6-34 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Data processing used to obtain the improved performance of a focused system
is a linear operation performed on a phase history that can be predicted for
each target in the antenna beam. This phase history or frequency history must
be recorded and operated on by the data processing. Unambiguous recording of
this signal can be obtained only if the system PRF is greater than twice the
bandwidth of the spectrum of the return from a typical target. This considera-
tion sets a lower bound for the radar pulse repetition frequency. A contrasting
upper bound is set by the range interval desired to be interrogated by the radar.
Ambiguity limitation becomes a problem when the lower limit of PRF is equal to,
or greater than, the prescribed upper limit of PRF. In such a case, the system
is said to be ambiguity limited and from ambiguity considerations a minimum
effective antenna area for the radar antenna is specified. The restricted min-
imum antenna area is a cost that must be borne if the advantages of a fully
focused synthetic aperture radar are to be obtained.
When the focused synthetic array radar is used, a marked improvement is
found in signal-to-noise performance over conventional radar. The most striking
improvement in this case is that the signal-to-noise ratio varies inversely as
the third power of range in a focused synthetic aperture system, while the sig-
nal-to-noise ratio varies inversely as the fourth power of range in a conven-
tional radar. This performance is obtained because the synthetic aperture system
essentially constructs an antenna for each range the radar is observing. The
length of this antenna is proportional to the range of the target. In effect,
the area of the transmitting antenna is fixed and the area of the receiving an-
tenna is proportional to the range of the target. The improved gain of this
synthetic receiving antenna yields an overall result in the signal-to-noise ex-
pression that the signal-to-noise power ratio varies inversely as the cube of
range rather than the conventional fourth power of range.
A system like a synthetic aperture radar is capable of collecting a fantas-
tic amount of data in a very short period of time. The problem of recording the
data for later data processing is a severe one. The most practical technique
available today is that of cathode-ray tube recording on photographic film.
Approximately 1000 independent events per linear inch of cathode-ray tube face
can be recorded with such a system. The consideration of resolution, recording
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
density capability, range interval, and usable width of film cictates the amount
of film that must be used in recording synthetic aperture radE.r data.
Optical data processing essentially transforms an unfocused image captured
in a radar space into a focused image in the optical data system. This tech-
nique of data processing has been utilized with synthetic aperture radars since
the earliest days of focused radar systems. The technique has proven quite
successful, and with optical data processing, it is possible to perform a task
that would require a tremendous amount of analog electronic equipment with a
relatively simple desk-top size data processing unit.
Several figures are attached to indicate the type of performance that can
be obtained from a focused, synthetic aperture radar. Figs. 6-8 and 6-9 are
photographs of unprocessed raw data film as recorded directly from a cathode-ray
tube-type recorder. A careful examination of these films will show in many
places the linearly frequency modulated target patterns which we discussed
earlier. The range direction on these pictures is in both cases across the
short dimension of the photograph. Observe that the linearly frequency modu-
lated target returns run normal to the range dimension and the essential data
processing that takes place in these systems is an operation in the longitudinal
direction. Figs. 6-10, 6-11 and 6-12, are various photographs made with synthet-
ic aperture radar. Examination of these pictures will indicate the order of fine
detail that can be obtained with the focused synthetic aperture radar system.
6.2 MOVING TARGET INDICATION (MTI)
Moving target indication radar covers many phases, including the detection
and recognition of moving targets and the determination of coordinates, velocity,
and direction. The presence of moving objects is an important clue in the
recognition of target complexes and in determining fluid military situations.
It is therefore desirable to have real-time airborne processing for MTI so that
it can serve as an alerting function, and can aid in selecting sensors for
limited spot coverage. An "ideal" MTI processor, for use with a high resolution
map display, would super-impose upon the display arrows indicating the magnitude
and direction of the target velocity vector at the correct position. A MTI
processor design should come as close to this ideal as possible with the input
data available.
6-:36 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
ph mare nsin. ;n[hetir aperture radar.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Fig. 6-11 - Photograph made using synthetic aperture radar.
6-38 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING.
There is a conflict in the optimization of radar parameters for the detec-
tion and resolution of both fixed and moving targets. MTI considerations alone
would dictate a forward looking radar to minimize doppler spread of the ground
return and thus would maximize the number of targets in the clear region. On the
other hand, to minimize the integration time required for a given resolution for
fixed targets, a side looking radar would be used. However, by making use of
the displaced phase center technique, it is possible to adequately cancel the
clutter even for a side looking radar, and still detect moving targets with
velocities much smaller than the maximum velocity spread over the real beam.
This MTI technique is compatible with the simultaneous generation of a high
resolution map, and this high resolution feature greatly reduces the area of
ground clutter with which the moving target competes. It is expected that this
type of MTI will have been thoroughly proven out by the post 1967.
The techniques used for a MTI processor are practically identical to those
used for the airborne processor for the high resolution radar map. In fact, it
is entirely feasible to design a processor which is compatible with both of
these functions. This electronic processor would be used as a relatively gross
resolution airborne monitor of the radar mapping performance. However, it may
prove desirable to have an electronic processor for the MTI function alone. For
the alerting function, MTI signals can be generated with azimuth and range posi-
tion data so that the area of interest can be mapped (by the radar and other
sensors) with fine resolution.
For example, the parameters for an alerting MTI radar system can be:
squint angle = 45 degrees
range resolution = 100 feet
azimuth resolution of clutter path = 100 feet
range measurement accuracy = 100 feet
The azimuth placement accuracy of a moving target depends upon the radial
velocity uncertainty, which for vehicles averages around 30 feet per second, and
for men, 3 feet per second. That is,
V 0
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
* 0
= angular uncertainty,
A V
V
= velocity uncertainty,
= velocity of aircraft and
0
= squint angle.
A 0
= 30 feet per seconds and 0
= 45 degrees,
A 0
= 0.014 radians for V = 3000 feet per seconds, and
0 0
= 0.035 radians for V = 1200 feet per seconds
The maximum angular uncertainty, however, will never be larger than the
beamwidth of the physical antenna and, by complex data, processing can be held
to a tenth of this beamwidth even for large velocity uncertainties. The azimuth
distance placement error derived from the above equations is plotted versus
range in Fig. 6-13.
The alerting time, defined, as the time between MTI detection and the cross-
ing of the target in the broadside direction is given by
T R cos
a V
zero = 10 nautical miles, 0 = 45 degrees
zero = 3000 feet per seconds, Ta = 143 seconds.
The limit on clutter cancellation ratio for the displaced phase center tech-
nique is set by equipment limitations and, based upon the performance of present
cancellation techniques, is estimated to be approximately -30 db. This is shown
for illustrative parameters in Fig. 6-14. Note that if there is considerable
ted for illustrative parameters in Fig. 6-14. Note that if there is considerable
internal motion of the ground clutter (for example, wind blown foliage), the
clutter cancellation ratio is considerably worse than -30 db. What these num-
bers mean in terms of actual detection of moving targets is illustrated in
Table 6-1. Obtaining fine resolution can be, by itself, an efficient clutter
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Nigh altitude
:3000 fps.
Low altitude
1200 fps.
Fig. 6-13 - Azimuth distance placement error vs range (MTI operation velocity
uncertainty = f 30 fps).
6-42 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Aircraft Velocity 1200 knots
Radar Wavelength 0.1 ft. (X-Band)
Displaced Phase Center
Clutter Cancellation
Distance Between Phase
Centers 6 ft.
Clutter Cancellation (Excluding
Internal Motion Effect)
I I I
SPECIAL HANDLING
V
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Table 6-1. Target Radar Cross-Sections
Actual Radar
Cross-Section
(sq. ft.)
Typical Radial
Velocity Due
To Own Motion
Effective
Cross-Section
After Clutter
Cancellation
(sq. ft.)
Vehicle
10-1000
15 kts
10-1000
Man
1-10 (est.)
2 kts
0.35-3.5
Grass 100' x 100' 20(1)
(
)
( )
0.02-2
1
25' x 25' 1.25
0-1 kt
0.001-0.12
Average 100'
x 100'
100(1)
0.1-10
Terrain 25'
x 25'
6.25(1)
0-1 kt(2)
0.006-0.62
Aircraft Velocity = 1200 kts
Radar Wavelength = 0.1 ft (X-Band)
Displaced Phase Center Clutter Cancellation
Distance Between Phase Centers = 6 ft
Maximum Cancellation (Equipment Limitations) _ -30 db
Combination of MTI with High Resolution Discrimination
(1)6 degree Grazing Angle: 77 = -17 db Grass
11 = -10 db Average Terrain
(2)The 1 knot velocity is for internal motion due to movement induced
by wind.
6-44 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
discrimination tool since the target of interest has only to compete with a very
small area of ground. Typical vehicles have a cross-section which is greater
than a 25 by 25-foot area of average terrain, and this ratio is increased an
order of magnitude by MTI techniques even with one knot internal motion. Detec-
tion of moving men, on the other hand, is not possible under all conditions when
there is internal motion of the ground clutter. Calculations indicate that, with
displaced phase center MTI in conjunction with high resolution on the order of
10-20 feet, it should be possible to detect moving men from a tactical reconnais-
sance aircraft under many conditions. It is reasonable to expect the develop-
ment and testing of techniques to do this by the post 1967.
It is feasible to use a cross-section (power return) threshold indicator in
conjunction with either the ground mapping or the moving target function of a
fine resolution synthetic array radar. This indicator would serve as an alert-
ing and location device. Since normal ground return exhibits a large fluctua-
tion of power return, the false alarm rate would probably be large unless the
information obtained by the device was coupled with other indicators such as
correspondence to an IR output.
It is, of course, not possible to choose specific system parameters for a
radar for a tactical reconnaissance aircraft without first defining the system
requirements fairly exactly. However, various tactical reconnaissance high
resolution radar system designs have been considered at Conductron, and the list
of parameters shown in Table 6-2 illustrate the principle feature of such a
radar for post 1967.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Table 6-2. Examples of SLR Parameters
Fine Resolution
System Type
Coherent SLR
Aircraft Velocity
Mach 3
Altitude(s) (ft.)
60,000
Maximum Range(s) (n. miles)
75
Range Interval Coverage(i) (n. miles)
Design Resolution(2) (ft.)
15 one side
30 total
Radiation Frequency 3)
X-Band
Average Power (watts)
500
Peak Power (KW)
360
Pulse Expansion Ratio (variable)
47-140
Pulse Length (Actual)
1.4 ?
Pulse Length (Effective)
Basic PRF (kc)
10-30 p
Physical Length (ft.)
Physical Height (ft.)
Bandwidth (mc) Video
33-100
Noise Temperature ?K
1000?K
System(4)
Estimated Volume (cu.
ft.)
15
Estimated Weight (lbs)
750
Estimated Power (watts)
4500
6-46 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w TW
SPECIAL HANDLING
Table 6-2 (Continued)
(1)
High altitude mode.
(2) Variable resolution, five feet, resolution for spot coverage to
50 miles.
For general mapping, low frequency option for foliage penetration
at short ranges.
(4) Including MTI and Pattern Recognition, but excluding displays and
Data Transmission.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
In reconnaissance, the use of optical scanning devices that operate in, the
infrared region of the spectrum has increased in recent years. Like photography,
infrared thermography produces a permanent record, but of data that cannot be
recorded by a camera working in the visible spectrum.
In effect, a camera observes the differences in light reflected from a
scene, and records this information graphically in a coordinate system which is
similar to that of the human eye. An infrared scanner, on the other hand, re-
cords differences in the emission of thermal energy or heat, and records this
information in graphic form after it has been converted to visible light. The
output of an infrared scanner is a thermograph, or heat map. As in photography,
one is concerned with the ground resolution of an infrared system, and also in
the temperature resolution, i.e., temperature differences.
Since the fundamental quantity being detected is differences in heat (emis-
sivity), the detecting element must be capable of sensing small differences in
temperature.
Since an infrared system, unlike a camera, is not dependent on reflected
sunlight, it will work as well at night as during the day time. It also has
better haze penetration than a camera, but it is not an all-weather device.
The interpretation of infrared thermographs employs many standard photo
interpretation techniques, but new techniques are also required. For example,
the sea presents large areas of uniform temperature and emissivity, and hence
is usually an ideal background for detection of ships. Sharp thermal differ-
ences are common at river outlets, Gulf Stream boundaries, underwater thermal
fronts, and where tidal currents flow over shallow banks. Ships' wakes are usu-
ally discernible because of their straight line character. Wakes of submarines
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 s
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
T M1
SPECIAL HANDLING
operating at 150 to 300 foot depths have been detected by infrared systems, par-
ticularly when the submarine performs a maneuver. The intensity of the infrared
radiation of ships is highly variable because of such factors as solar heating,
wind velocity, air and water temperature differences, the mode of propulsion,
and exhaust product disposal. (For example, in one test, a hot spot on-the deck
appeared on some infrared recordings but not on others. The commanding officer
commented that it was the habit of the ship's baker to bake after the evening
movie.)
One degree of temperature difference permits satisfactory reconnaissance of
natural land scenes, even for such unusual missions as ice thickness studies and
ice crevasse detection. Solar heating, air temperature and wind histories deter-
mine contrast. Contrast is frequently reversed at night because water, for ex-
ample, does not cool as rapidly as land surfaces. A solid cloud cover overhead
reduces contrast because it affects cooling and allows the ground to reflect the
thermal radiation from relatively warm clouds rather than from the cold sky.
Clouds between an infrared system and the ground block the radiation, and occa-
sional incipient clouds or thermal turbulence will mar thermal maps.
Manmade objects seldom match the thermal characteristics of natural objects
and hence are easily detected. Furthermore, human activity requires the expen-
diture of energy which usually dissipates into thermal energy. Tanks, cars and
even men are detectable, but the detection of the latter usually requires prior
knowledge. The cooking of food provides thermal sources which can sometimes be
detected even under dense foliage because of holes in the foliage. Multiple
coverage (overlap or repeat scans) can significantly increase the probability of
detection by increasing the probability of "source-hole" alignment.
The trade-offs considered in designing infrared systems are somewhat anal-
ogous to those considered in designing a photographic system. The designer is
concerned with resolution (both angular and temperature), angular coverage, and
the V/h rate which, when combined with resolution and angular coverage, deter-
mines the system bandwidth or-data rate.
In a photographic system, the resolution is fundamentally limited by the
Rayleigh limit, i.e., the system is aperture limited. In a given set of cir-
cumstances, however, image blur may impose a more severe limit than aperture
(or lens-film resolution).
7-2 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
In infrared system design, where a large aperture is also desired, limita-
tions are not imposed by the Rayleigh limit but by a photon or energy limit.
All objects, manmade or natural, emit energy if they are at any temperature
above absolute zero. As the temperature increases, the amount of energy emitted
increases but the wavelength of the radiation decreases. Hence, an object at
5000?K emits radiation over a wide band of wavelengths, from the far infrared
(around 80 microns) up to well beyond the visible spectrum. At this temperature,
the radiation is highest at the red end of the visible spectrum. However, it
radiates a great deal of energy in all wavelengths of the visible spectrum and
thus would appear to be white hot. A cooler object of 1000?K would have its
peak radiation around 2 1/2 microns. However, it would emit enough radiation
in the visible portion of the spectrum to appear to be red hot. In other words,
it would be visible to the human eye. As the temperature of the emitter de-
creases, the peak wavelength shifts toward longer wavelengths. Hence, a room
temperature object that is 293?K or 20?C peaks at about 10 microns, which is
well beyond the visible spectrum. For reconnaissance use, many of the objects
of interest are emitting most of the radiation at rather long wavelengths. A
room temperature object is emitting a small but detectable amount of radiation
in the 4-micron region, but as noted, it peaks in the 10-micron region. Hence,
if one is interested in relatively cool objects, that is, objects which are a
few degrees above the ambient temperature, it is desirable to use detectors
which are sensitive to the 10-micron region of the spectrum.
The atmosphere imposes an additional constraint. The atmosphere is reason-
ably opaque to radiation in the 4 to 8-micron region of the spectrum. Fairly
good atmospheric windows exist between 2 to 4 microns and between 8 to 14 microns.
Working in the infrared region of the spectrum imposes a problem in selec-
ting window materials. For one thing the window itself will heat up if used in
a high performance aircraft and appear as a heat source to the infrared system.
For this reason, an opening in the bottom of the aircraft should serve as the
"window".
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 w
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
7.2 DETECTION METHODS
The method of detecting thermal radiation is an important factor in infrared
system design, and is discussed briefly here since it affects the optical system.
Photographic films are not satisfactory since the radiation at room temperature
would fog the film at the instant of manufacture, unless the film were manufac-
tured at temperatures below one degree Kelvin, which, of course, is impractical.
Detection devices which permit imaging a scene on a detector other than film
have been evolved. These include the Baird Atomic evaporagraph and the
Westinghouse thermicon (thermal vidicon). However, neither of these devices are
capable of handling an adequate information rate with sufficient sensitivity to
provide the tenth of a degree to one degree temperature resolution required for
aerial reconnaissance. It seems unlikely that this sort of a detection method
will be developed to an adequate degree to meet the reconnaissance requirements
of the next generation of equipment.
The only feasible method of detecting thermal radiation is to scan the scene
with single or multiple element detectors and reconstruct the scene in a manner
similar to the early television technique of the Nipkow disc. Usually the detec-
tor and imaging optics are held stationary and a rotating flat mirror moves the
line of sight of the instrument. In aerial reconnaissance, scanning in the sec-
ond direction is accomplished by the aircraft motion, while in ground-based
reconnaissance systems the rotating mirror is slowly tilted. The resultant elec-
trical signal from the detector is used to modulate a light source which, in
turn, is imaged on standard photographic film with a scanning system similar to
the infrared optics.
Table 7-1 summarizes pertinent information on several detectors which have
been used successfully in infrared systems. Ideally, all thermal radiation
should be used for maximum sensitivity. However, as noted in the table, differ-
ent detectors respond in different regions of the spectrum, and their information
rate varies. Also, working in the longer wavelengths increases the cooling re-
quirements of the detector, since the detector must be cooled below a point where
its own noise affects performance. Presently, the germanium mercury detector
appears to offer the best combination of characteristics for an infrared system.
7-4 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
Vw
SPECIAL HANDLING
rH
cd
LO
O cam!) ? rH-?I CJ co
Q) r-I
14
bO
0 0 0 0 0
oo C') oo`o 00
oo LA
4-I
u
a)
y
G x
0
.u O
oO
P
G
a)
Cl)
O
r-~
O O
H T-4
b
`Vp
DC 5C
O
o
u
P 04
rH-I
C9 00
~- O
0 -W
U
cd
P;
P H
O cd
" ?rl
u P
0) Q)
41 4-J
Q)
H CO
?rl
a)
bO
0
u
m
o
x
0
cd
c
4
1
a)
a)
P4
H
H
0
0
ra
a)
r4
cd
H
bO
O
O
O
u
It
O
CO
cV
cq
ca
u
O
O
to
H
cd
~d
a)
tO
H
LCD
0
O
ca
u
?
tom"
cli
'Ili
LO
'14
U'~
,4
CV
O
H
r-I
a)
I
I
I
I
I
4-I
4-1
O
(D
IM
O
O
10
r-I
CO
er
CO
cL
SPECIAL , HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
9C 0
I0 U
E--
r+
0.
0
L
V
0) L
C o
C u
m 0
u 7
un
c C
o w
~. u
U 10
u
O V)
F+
7-6 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
1W
SPECIAL HANDLING
It is sensitive in the 8 to 14-micron region of the spectrum, and its informa-
tion rate is high. It has a short time constraint - something less than a micro-
second - and the signal-to-noise ratio is excellent.
Within the next few years the performance of infrared detectors should be
within a factor of three of the theoretical limits imposed by the random arrival
of photons. (This condition is referred to as BLIP - Background Limited Infrared
detection). The average photon rate for a 300?K black body is given in Table 7-1
as well as the rate difference for one degree Kelvin temperature difference.
The maximum theoretical signal-to-noise ratio is given for a typical reconnais-
sance requirement and system. The system is assumed to cover a field of view of
one milliradian on a side, a million such viewing elements per second, a one
degree temperature difference, a radiation collecting aperture of 100 square
centimeters and a single detector.
System performance is usually measured in the number of elements viewed per
second, system resolution (angular size of the field of view), and temperature
difference required for a signal-to-noise ratio of one NETD (Noise Equivalent
Temperature Difference).
Fig. 7-1 shows the essential functional elements of the system. The scene is
scanned and imaged on a cooled infrared detector. The electrical signal is
amplified and converted to an equivalent intensity light signal which is imaged
on to film via roll-compensated scanning optics. The film is moved forward at a
speed proportional to V/h. Scanner position, roll compensation, V/h, as well as
the electrical signal from the infrared detector, must be telemetered in order
to reconstruct the image on the ground. By setting a threshold, intense infra-
red signals can be converted to an alerting signal for recording on film or acti-
vating other sensors.
The design of the scanning optics is determined by the angular coverage re-
quired and the V/h limits. As V/h increases, rotational speeds must be increased
proportionally. Generally, flat optical surfaces are used because they can best
be mounted without excessive image distortion. Prisms can also be used. How-
ever, the choice of material is limited, particularly when working in the 8 to
14-micron region.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
In designing the imaging optics, reflective systems are generally used since
they cover the wide spectral region without severe chromatic aberration. As
indicated previously, the imaging requirements generally do not approach the
diffraction limit or Rayleigh limit. The twenty-fold increases in wavelength
over those utilized for photographic systems, make the optical tolerances less
severe. Since the radiation collecting area requirements are large, the dif-
fraction limit is more significant. In some designs the detector itself is
placed in the image plane (where the resultant image size, a function of scale -
focal length divided by altitude - is compatible with the detector area). In
most cases the imaging optics form an image which is transferred by a field lens
to the detector. In other words, the field lens images the entrance aperture on
the infrared detector. The field lens is usually designed to work in the spec-
tral region of the particular detector. Cone condensers can also be used in
place of the field lens.
The spectral filters are usually placed close to the detector to optimize
image quality by eliminating unwanted radiation from the scene. The filter may
be an integral part. of the detector assembly.
Infrared detectors suitable for reconnaissance must have short time con-
stants to handle the high information rates required. The bit rate or data rate
of an infrared system is given by the formula Dr = VS/h 2, where Dr is the data
rate in bits per second, V is the ground speed, h is altitude, S is the cross-
track angle in radians, and 0 is the angular resolution in radians. At a V/h of
two radians per second (approximately Mach I at 500 feet), a resolution require-
ment of one milliradian (corresponding to six-inch ground resolution at 500 feet
at nadir), and a scan angle of 120 degrees or roughly two radians would result
in a data rate of four million bits per second or four megacycles. The data
rate is directly proportional to V/h and to scan or cross-track angle, but in-
creases as the square of the angular resolution. Hence, a requirement for a
1/2 milliradian resolution with the same set of parameters previously given
would require a 16 megacycle width. Thus, it can be seen that the response time
or time constant of an infrared detector must be very short -less than a micro-
second if the number of detectors in the system is to be kept in reasonable
bounds. A single detector system where the detector has a time constant of less
than a microsecond could handle a one megacycle data rate. The signal-to-noise
7-8 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
w
SPECIAL HANDLING
ratio tends to improve if the detector is not pushed too hard in terms of res-
ponse time. It is generally agreed that present single detectors should not be
pushed beyond a one megacycle data rate, and even this is considered a rather
severe rate. Hence, if one considers the problem of infrared reconnaissance at
V/h rates greater than two radians per second with resolution requirements of
1/2 milliradian or better, multiple detector systems are required.
At present, infrared detectors used for high information rates are either
intrinsic or extrinsic photo detectors. Chemically deposited materials such as
PbS and PbSe exhibit time constants which are too long and further require con-
siderable electronic compensation for data rates. Intrinsic semiconductor crys-
tals such as InAs and InSb have been developed in photodiode form for wavelengths
to seven microns. Longer wavelength response is obtained with extrinsic detec-
tors such as copper or mercury-doped germanium. Since these detectors exhibit
low absorption, integration chambers are required to achieve multiple paths
through the crystal. At the present state of the art, these detectors are the
preferred choice. However, intrinsic detectors such as HgTe are being investi-
gated and may be available in the next few years.
Cooling to cryogenic temperatures is required for most detector operation.
Thermoelectric cooling can be utilized for temperatures down to 200?K. Gas
cooling is employed below 200?K by liquefying mechanisms. Liquid nitrogen from
storage containers, as well as cryostats utilizing high pressure nitrogen to
achieve 77?K, are used for this purpose. Similarly, neon, hydrogen and helium
are used for lower temperatures. The use of cryogenic solids such as methane
has also been explored.
In addition to cooling, infrared detectors require low noise preamplifiers
because the signal levels are near the Johnson noise. Such preamplifiers are
well within current technology.
Due to detector time constant and capacitance associated with high resis-
tance, high frequency boost is sometimes required. If multiple detectors are
used, some form of keyed or subliminal agc is required to equalize response.
Recent systems have employed a threshold to perform an alerting function such
as marking the edge of the film. On certain missions such as submarine wake
detection, temperature gradient is more significant and some degree of low fre-
quency suppression is used.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
f
SPECIAL HANDLING
Several display methods are used with IR systems. An "A" scope display is
sometimes utilized. Because information from previous scans is needed for a
more adequate understanding, film recording and rapid development are a necessity.
Recording techniques other than film may be used for "instantaneous" display and
may be satisfactorily developed by 1967. Updating a video display requires con-
siderable equipment. and has an extremely limited time span compared to a "hard
copy" process.
Film recording is usually accomplished with a modulated light source, such
as a glow tube, and scanning optics that are mechanically coupled to the infra-
red optics through a roll compensation mechanism. The film speed is controlled
by the V/h detector. Electronic scanning and compensation through the use of a
CRT, while allowing less mechanical parts, introduces more stringent linearity,
resolution, dynamic range and roll, pitch, and yaw compensation requirements.
Magnetic tape recording is presently used in certain applications, but re-
quires more ground equipment and better equipment accuracy to allow for signal
degradation during recording and playback. For high data rates, video recorders
beyond the present state of the art would be required.
As with other sensors, the design of an operational infrared system is also
limited to space available in the aircraft. A typical system would scan over
120 degrees. A 180-degree system is feasible; however, information gathered
beyond 120 or 140 degrees is of questionable value.
Angular resolution of less than a milliradian and temperature resolution of
0.1 degrees Kelvin are feasible. Such a system would weigh about 200 pounds and
require about 1000 watts of power (400 cps three-phase). The overall size would
be about 1.5 by 1.5 by 3 feet.
7.4 INFRARED RECONNAISSANCE SYSTEM DEVELOPMENT
The development of infrared scanners of the type envisioned for the multi-
sensor mission is less than a decade old. From 1950 to 1960, the infrared recon-
naissance system evolved very slowly as a low altitude, low resolution, low V/h
device, with expanding capability only in terms of spectral coverage at longer
wavelengths. Resolutions of several milliradians, temperature sensitivities of
1?C, and V/h capabilities of 0.050 radians per second were typical requirements.
7-10 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
NW
Operation in the lead sulphide spectral band (1 to 3 microns) was succeeded by
use of cooled lead selenide detectors in the 3 to 5-micron band and finally by
copper-doped germanium and mercury-doped germanium detectors in the 8 to 14-
micron band.
Until several years ago, the angular resolution requirement remained in
the range 2 to 3 milliradians although temperature sensitivities as low as 0.1?C
(with entrance aperture completely filled) had been reached. Indium antimonide
and lead telluride detectors were developed allowing operation in the 4.5 to
5.5-micron window. The AN/UAS-5, AN/AAR-9, and AN/AAD-2 scanners are represen-
tative systems of this era.
Systems described within the past year or so (the AN/AAS-18 and AN/AAS-10)
have resolutions exceeding 1 milliradian with a V/h capability to over 2 radians
seconds and noise equivalent temperatures of 0.25?C. Many useful and high qual-
ity examples of terrain imagery have been produced by the existing systems when
operating at the lower altitude.
However, if one considers a reconnaissance mission with a contemporary
high-speed high-altitude aircraft at an altitude of 50,000 feet, an angular res-
olution of 1 milliradian would resolve ground structure of the order of 50 feet.
Imagery of objects whose size is much less than the width of an airstrip would
be seriously degraded. To achieve ground resolution comparable to that obtained
with a 1 milliradian system at 5000 feet, the instantaneous angular resolution
requirement becomes 0.1 milliradian. Because of the inverse squared dependence
of noise equivalent temperature difference on resolution, the sensitivity of the
system would have to be increased by a factor of 100 to preserve temperature
resolution comparable to that obtained at the lower altitude. Otherwise, a
noise equivalent temperature difference of 0.3?C would become 30?C, and terrain
imagery would be impossible because the temperature differences associated with
terrain features are in the 1 to 10?C range. In addition, the increased data-
gathering rates imposed by the higher resolution may impose requirements that
will render the past generation systems unsuitable for generalized tactical
reconnaissance.
Faced with these technical requirements, it appears mandatory that single
channel detection systems, which have been satisfactory at the lower altitudes
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
and speeds, be replaced with multi-element detector arrays followed with multi-
channel electronics in order to cope with the requirements of sensitivity, data
rate, and scanning speed. Multi-element sensors having of the order of ten de-
tectors have been designed and flown with excellent results.
In summary, infrared surveillance systems have not been widely used in op-
erational aircraft reconnaissance. The traditional single-detector channel sys-
tems have not been able to satisfy the requirements of resolution and bandwidth
imposed by present high speed aircraft operating under generalized mission-
profile requirements. However, recent advances in the development and fabrica-
tion of multiple detector arrays offer a potential for infrared reconnaissance
systems that has not been previously realized. The parametric analysis in the
next section establishes the relations that would be used in setting forth sys-
tem specifications.
7.5 INFRARED PARAMETRIC ANALYSIS
This analysis is based on a panoramic scan at right angles to the vehicle
track by an array of detectors. The output format is of the strip map type. In
actuality, the detector array is stationary and the scan action is produced by a
mechanically rotating reflector, assumed to be a single element mirror for the
present purpose. This mode of scan is also called the "sweep broom" type.
7.5.1 Diffraction Blur Circle
The minimum diameter D of the entrance aperture of the optical system
necessary to provide the desired angular resolution is
where 0d is the angular diameter of the central maximum of the diffraction pat-
tern and 'A is the wavelength. This is the least requirement on aperture irre-
spective of sensitivity requirements. At a wavelength of 10 microns, and at an
angular resolution of 0.1 milliradian, the least optical aperture is about 10
inches.
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
7.5.2 Minimum System Focal Length
On the basis of the minimum detector size A d which can be fabricated and
the desired angular resolution A 0, the minimum focal length f is found from
f d
= 0
Doped germanium detectors suitable for use in the 8 to 14-micron atmospheric
window can be made as small as 0.004 inch in size.
7.5.3 Noise Equivalent Bandwidth
The noise equivalent bandwidth A f, or the single channel bandwidth with a
single-element scanning mirror is given by
f IT v/h
(1-a) n (,L 0)2
V = velocity of the aircraft,
h = aircraft altitude,
n = number of detector channels,
L 0 = angular resolution of a single detector, and
a = fractional overlap of successive sweeps.
0 f =
IL 0
1-a) n h i 0
d
= 2
~
2 =t ' Q t
, an
T=
T
V
where
t =
detector dwell time for a point source
and
T =
scan period of rotation during which time the aircraft advances
(1-a) n h A 0.
If the fractional overlap can be neglected,
A f = IT V h
n (A 0)2
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
Doped germanium detectors have maximum useful bandwidths of the order of 1 mega-
cycle per second. To conform to this requirement, it is necessary to increase
the number of channels as V/h increases with high speed aircraft or as the angu-
lar resolution is decreased for high altitude operation. With single channel
systems, A f may become prohibitively high.
7.5.4 Total System Bandwidth
The total system bandwidth 6 fT is increased by the number of detector
channels operating in parallel according to
A fT = n n f .
7.5.5 Average Information Data Rate
This is defined here as the number of scanned resolution elements swept in
a period of rotation and it does not comprise the out-of-scan elements during
retrace. It is given by
V S
DR =
h (A 8) 2 (1-a)
where S is the cross-track angle, and the other symbols are as defined above.
It includes the redundant elements scanned twice because of overlapping of suc-
cessive sweeps.
The number of once-scanned resolution elements is found by application of
the factor 1-a, where o! is the overlap, and the information rate for these is
given by
DR =
o h (A 0)2
this is the same rate as for no overlap.
7.5.6 Noise Equivalent Power
The noise equivalent power (NEP) of the detector is given by
AZ f
NEP = D*
7-14 SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
Nkw %W
SPECIAL HANDLING
where A is the detector area in cm2, 1 f is the noise equivalent bandwidth, and
D* is the photodetector figure of merit normalized to unit frequency bandwidth
and unit area. For doped germanium detectors, the peak D* may approach
loi i cm cps 1/2
watt
7.5.7 Noise Equivalent Temperature Difference
The noise equivalent temperature difference (NETD or A TN) is an important
parameter describing the performance of an infrared surveillance system. It
represents that temperature difference of a blackbody radiator with respect to
the ambient reference temperature and which provides for a signal-to-noise ratio
of unity. The radiator completely fills the field-of-view of a single detector.
0 TN is found as follows.
NEP = A P = a
a TT A TN ,
since two adjacent fields-of-view are at different temperatures 0 TN apart and
the change in power at the detector is L P. But,
P = eo eE 7r 42 (0 9) 2 N
eo, eE = optical, electronic efficiencies,
D = diameter of the entrance aperture (cm),
0 = the angular resolution,
N = the apparent radiance (watts ).
cm2 - sr
The apparent radiance is found by weighing the Planck blackbody radiance
function W/7r (T, T) where T is absolute temperature and ?\ is the wavelength, by
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
SPECIAL HANDLING
the normalized spectral response of the detector D*/D*max and by T (~,), the com-
bined atmospheric and filter transmittance. Integration over the spectral in-
terval gives
?
D* T (T) n (T, h) dA ('atts
max cm -sr
Further, introducing the bandwidth
f 7r V/h
n (A B)2
and the detector area
A = ( f A 8) 2
and by doing some algebraic manipulation, it is found that the NETD is
ATN=
(A d)2
4 f n V/h
n
A2
D2 eo eE
I
'Al
aW
TT
The integral in the denominator has been evaluated at a background reference
temperature of 288?K in the 8 to 14-micron region with doped germanium to be
1/2
3 x 106 (cm cis
?K
7.5.8 "Push Broom" Scan
In this type of scan, there is no rotation of the detector array. The num-
ber of detectors, n, is given by the ratio of the cross-track angle S to the
angular resolution A 0. The downward-looking field-of-view, in conjunction with
the aircraft motion, maps a strip along the vehicle track. This type of scan is
more suited to low altitude operation where the requirement on angular resolution
SPECIAL HANDLING
Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
 Approved For Release 2009/07/08: CIA-RDP67B00657R000300210001-4
W
SPECIAL HANDLING
is less severe. Otherwise, in a high altitude aircraft the number of detectors
can be prohibitively large.
It can be shown that the expression for NETD in the section above for the
"sweep broom" scan is also applicable to the "push broom" scan if the quantity Tr
under the radical is replaced by S/2.
7.5.9 Signal-to-Noise Ratio, "Sweep Broom" Scan
The determination of target signal follows along the methods used in the
analysis of NETD. Analogous to A TN, L 0, and NEP, we find the target temper-
ature difference 0 TS, the angular subtense of the target at the entrance aper-
ture A w, and the signal power S. Following through the analysis, it can be
shown that the signal-to-noise ratio is
2 A T
S/N = A w 0 TS where A