(Sanitized) COMPUTER DISPLAY

• Document Type: CREST
• Collection: NGA Records (Formerly NIMA)
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78B04747A001100030003-4
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 3
• Document Creation Date: December 28, 2016
• Document Release Date: May 6, 2003
• Sequence Number: 3
• Content Type: MFR

Approved Foy-Release 2b0/051
25X1R5X1A
25X1A
MEMORANDUM FOR THE RECORD
1.. In response to a request from
ESD/EI~ was asked to comment on the
micro-densitometer display equipment.
2. On 197ebruary 1969,
ESD/EL, attended a briefin;; in
Chief/ESD,
office ;_iven by
The purpose of this
briefing was to display the results of a high sped raster scan hard copy
printer and ddiscuss some of its properties. After the -criefin ,
For his information,
to comment in writing on what he had seen and heard.
C . i+In nnn~oon~~~~nn AAAAA,AAA9 A.
25X1A
.25X1A
Te-a -T-7 _
LfG ,
5
~
,
rQ
was supplied a copy of an MFTR dated
11 February 1969, which expressed some of
feelings on the
same subject when he got essentially the same briefing on 6 February 1969.
3. The display device produced a raster scanned, ink.-on-paper print
and was billed as one compo.ent of total imadcaiing system. The printer
has a good grey scale range, a raster fine enough --er=as to ~e barely dis-
cernable when viewed at a distance of two feet or so, and a large print
capatility (approximately briefing-ward size. Through multiple cycling
full color or false color prints can be produced.
Declass Review by NIMA/DOD
Approved For Release 2003/05/15: CIA-RDP7 4'7-? 7AOO41 0030003-4
Chief/ESD, arid
25X1A
25X1A
25X1A
25X1A
25X1A
25X1A
25X1A
 Approved For Release 2003/05/15 '?CIA-RDP78B04747Kb011000i0003-4
4. The overall. systc-rn into which this device interfaces contains
a modified scanning, microdeositometer, an IBM 360/40 digital com-
puter, and the aforementioned display unit. The full input/output cycle
starts with raster-samplG'd density data from imagery on the micro-
densitometer. The sampling rate sampling aperture, and scan length are
independently adjustable, thus permitting speed, resolution,Amatrix-size
adjustment. This data is then fed into the computer where it is cof6rt3ected
for input scan imperfections, manipulated, formated, and printed out using
the display device. The display device is controlled by the computer.
5. The manipulation capaLA-lities of the computer potentially run
the gamut of what can be done to either enhance or restore the imagery.
The typical enhancement techniques of density clipping, contrast building,
or making non-linearity colY,t ctions can all be done using large data
matrices as demonstrated on the l9th; however, to apply restoration techniques
for defocus, image motion, atmospheric problems, or other system transfer/
cot-; sections, the use of Fourier Transform method; would be required. On
the 36O/40, this would limit the working matrix to a 128 X 128 arm which
would yield only twice the format size of the system at `?
equivalent output raster frequencies.
6. The prints displayed on the l9th demonstrated that
has built what must be the worlds most expensi\/eenlarger. The grey scale
manipulation which was done could be approximated through darkroom pro-
cedures and the grain sup-pressior: could be accompolished through careful
defocusin, of a conventio al enlarging system. Techniques requiring the
use of Fourier Transforms were not demonstrated, and indeed could not be
for the format size displayed, unless 128 X 128 matrices were extracted
25X1
25X1A
frorA `dtForR416ai6d2 /68 'adAyRD 66-eMM66i1i6b 6 4form to
-2-
 Approved For Release 2003/05/15: CIA-RDP78BO4747A001100030003-4
reconstruct a printout. 't'his would be very time consuming at rest.
7. If a non-silver photographic, computer co:rtrolled enlarger is
what the center needs, then the ystem should certainly be considered.2 5X1A
If on the other hand, an Image Restoration Facility is acquired, this system
should also be considered; however, a final decision should be withheld pending
the esta'Uhm ent of certain operating; parameters for_.dgital ima
restoration techniques. At the present time, for example, the extraction
of information throu,h the use of Fourier transform image restoration
techniques has not,: demonstrated for cases where the required information
is in the 50-150 line/per millimeter frequency range.
It is the opinion
of this writer thatLthe committment of funds of this magnitude (1.6 million
should not be considered until this capability can be demonstrated. This,
by the way is the primary o-fic.tal goal of the
D. I. R. program. 25X1 A
Approved For Release 2003/05/15 : CIA "04747A001100030003-4