REVIEW OF SOVIET PHOTOGRAMMETRIC PROCEDURES
Document Type:
Collection:
Document Number (FOIA) /ESDN (CREST):
CIA-RDP79-01083A000100040001-4
Release Decision:
RIPPUB
Original Classification:
S
Document Page Count:
82
Document Creation Date:
November 9, 2016
Document Release Date:
August 18, 1998
Sequence Number:
1
Case Number:
Publication Date:
May 1, 1954
Content Type:
REPORT
File:
Attachment | Size |
---|---|
![]() | 3.37 MB |
Body:
SECRETApproved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
Copy No.
PROVISIONAL INTELLIGENCE REPORT
REVIEW OF SOVIET PHOTOGRAMMETRIC
PROCEDURES
CIA/RR-ER-4
May 1954
25X1A5a1
CENTRAL INTELLIGENCE AGENCY
OFFICE OF RESEARCH AND REPORTS
DOCUMENT NO.
NO CHANGE IN CLASS. ~,.r.., m; W
u DECLASSIFIED SECRET
Approved For " 0~A P79-01083A000100040001-4
AUTH: HR 70.2
DATE:/~~l-q.REVIEWER: ji9360
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
WARNING
This material contains information affecting
the National Defense of the United States
within the meaning of the espionage laws,
Title 18, USC, Secs. 793 and 794, the trans-
mission or revelation of which in any manner
to an unauthorized person is prohibited by law.
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
mdagam
PROVISIONAL INTELLIGENCE REPORT
REVIEW OF SOVIET PIJ)TOGRAMMETRIC PROCEDURES
CIA/RR-ER-4
CENTRAL INTELLIGENCE AGENCY
Office of Research and Reports
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
FOREWORD
This report is one of a series prepared for an external research
project entitled "The Problem of Soviet Capabilities in Geodesy and
Cartography," which was sponsored by the CIA as an element of the research
program of the Geography Division, Office of Research and Reports. The
25X1A5a1
Inadequate knowledge concerning the status of Soviet developments
in geodesy, photogramanetry, and cartography has been recognized as an
intelligence deficiency, and intelligence reports in these fields are
extremely few in number and limited in topical coverage. For these
reasons, the project was initiated to assess Soviet capabilities on the
basis of a systematic study of all available published information on
Soviet developments in geodesy, geodetic gravimetry, geodetic astronomy,
geodetic and photogrammetric instrumentation, and cartography. The
resulting reports are derived almost entirely from an exhaustive search
for and an analysis of published Soviet scientific source materials.
The reports of the Project are designed not only to provide pro-
visional information on the current status of Soviet capabilities in
surveying and mapping but also to serve as a datum for a continuing
Approved For Release 1999/09/02 : CW~DW79-01083A000100040001-4
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
program of collection and research that will in the future provide
increasingly accurate and timely intelligence. The opinions and con-
clusions presented in -ER-4, therefore, do not represent final CIA, evalu-
ation of Soviet capabilities in surveying and mapping. Comments by
users of the reports are solicited by the Geography Division.
This report, which is one of four on Soviet photogram+netry, is of.
greatest value when read in conjunction with the report on Soviet
Geodetic and Photogrammetric Instrumentation, CIA/RR-ER-S. The descrip-
tion and analysis of photogrammetric procedures in ER-4 is based on
information obtained from Soviet source materials, dating from 1931
through 1951. For readers requiring additional details or information
on other aspects of Soviet photogrammetry, an extensive bibliography
has been compiled and is available on request, through appropriate chan-
nels, as CIA/RR-ER-7, A List of References to Soviet Photogrammetric
Literature. For further information concerning Soviet instruments, a
list of references has been compiled and is also available on request
as CIA/RR-ER-6., Additional References to Soviet Diagrams, Sketches, and
Photographs of Soviet Geodetic and Photogrammetri.c Instruments.
Part I of this report is a summary of the development of Soviet
photogrammetry. Part II provides a general description and appraisal
of methods and procedures. Technical appendixes give more detailed des-
criptions and appraisals of seven Soviet books that have been selected
as representative basic works in Soviet photogrammetry.
Since the analyses are based entirely on textual materials, they
are not definitive. Such an evaluation can be obtained only through
a critical analysis of Soviet instruments and an extended comparison
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
of Soviet maps with aerial photographs. Such depth of analysis is not
currently possible because of the unavailability of Soviet instruments
and Soviet aerial photography used in the compilation of the Soviet maps
now in this country.
Publications produced under terms of this contract and issued to
date are s
Geodetic Gravimetry in the USSR, CIA/RR-ER-1j, 18 October 1951?
SECRET.
Deformation of the Crust of the Earth and Terrestrial Magnetism,
CIA/R -ER- , 1 October 19 1. SECRET.
General Critique of Soviet Gravimetric Data with an Annotated
Bibliography, CIA/RR-ER-3., April . SECRET.
Selected Bibliography of Soviet Studies in the Field of Cosmic
Ras, CIA/SI 7b-54., 2 March . Cover and foreword,, SECRET; io-
graphy, Unclassified.
Review of Soviet Photogrammetric Procedures, CIA/RR-ER-14, May,
1954. SECRET.
Soviet Geodetic and Photogrammetric Instrumentation, C Lk/RR-ER-5.,
15 April 1954. SECRET.
Additional References to Soviet Diagrams., Sketches, and Photographs
of Soviet Geodetic and Photogrammetr c Instruments, CIA ER ,-6 May,
17 94. SECRET.
A List of References to Soviet Photo rammetric Literature, CIA/RR-
ER-7, May, 1954-9 .
'Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
5X1A5a1
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/d?A-RDP79-01083A000100040001-4
CONTENTS
Page
Excerpt from speech by Academician A. N. Nesmeyanov . . . . , i
Summary and Conclusions , . . . ii
I. Development of Soviet Photogrammetry . . . . . . . . . . 1
A. Introduction . . . . . . . . . . . . . . . . . , . 1
B. Historical Development of Photogrammetry in the USSR 5
II. Soviet Mapping Scales and Procedures . . . . . . . 9
Appendixes
25X1A5a1
Appendix I. Comments and Discussions on the
Soviet book entitled: Nastavleniye o topografi-
cheskoy s"yemke v masshtabe 1:10 . Cast II.
ogramme richeskiye rabot (Instructions
for
Topographic Surveys on a 1:100.,000 Scale, Part II.
Photogrammetric Operations) Second Edition, GUGK,
1950. . . . . . . . . . . . . . . . . . . . . 13
25X1A5a1
Appendix II. Review of Foto rammetrira (Photo-
grammetry by N. N. VejgkTA %, 5. . . . . 18
Appendix III. General Discussion of
Geodezi a, Tom IX, 19 9, Chapter 1. (Small Scale
g
Aerial Photo raphy) . . . . . . . . . . . . . 28
25X1A5a1
Appendix IV. Discussion of Ste reofoto-
graminetri Ste reophotogrammetry y A .
Sk r dov25 5e 1' ? . . . 39
Appendix V. Report on Soviet Photogram-
metric Ins ruments as described in Foto ram-
metricheski Pribo i Inst rumentove en ye
(Photogrammetric Apparatus and instrumenta-
tion) by F. V. Drobyshev, 1951. . . . . . 46
Appendix VI. Comparison of Accura2 X1A ping Processes in
the USSR and USA 56
Appendix VII. The Straight Line Method of the Soviets . . . 59
Appendix VIII. Bibliography . . . . . . . . . . . . . . . . 70
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECFM
Approved For Release 1999/09/02 3 RDP79-01083A000100040001-4
..."The task of maximum use of natural resources, set forth by the
19th Congress of the Party, is connected with the execution of large-
scale and complex investigations, the basic aim of which is the improve-
ment of geographic distribution of industrial establishments, which has
in view as far as possible bringing industry close to the sources of raw
material and fuel.
This directive demands of science a multi-sided study and mastery of
the vast expanses of our East and South. Before the Academy of Sciences
is laid the task -- on the basis of broadly developed geological, geographic,
and economic investigations, carried on together with various offices and
ministries, -- to furnish the scientific foundation of the proposed devel-
opment of the national economy of a number of eastern and sourthern regions
of the Soviet Union.
The use of new methods of aerial investigation for territorial (air)
route research of a locality has basically change d the tempo of investi-
tine oueratione"...
Excerpt from a speech by Academician,
A. N. Nesmeyanov
printed in
Vestnik Akad.emii Nauk SSSR
Vyp. 3 (March), 1953
Approved For Release 1999/09/02 : CE RDP79-01083A000100040001-4
Approved For Release 1991b2 : CIA-RDP79-01083A000100040001-4
The Soviets appear to have made several original contributions in the
field of photogrammetry. These are probably, in the order of their impor-
tance, as follows:
1. The development of super wide-angle camera lenses.
2. The development of efficient small-scale mapping systems
based on the Odifferentiated" method, though the method
itself is not an entirely original concept of the Russians.
3. The concept and design of Drobyshev's stereometers.
4. The "straight-line" method of Romanovskiy.
Major trends in Soviet photogrammetrical methods as commented on more
fully later would seem to be
1. A recognition of the importance of the optics, and a
development along thoselines possibly exceeding our own,
,if the reports-on their lenses are true.
2. The tendency to make the corrections for orientation
mechanically, rather than by homolog or optical means.
3. The use of numerical computations where we -tend to
compute by analog, or not to -compute explicitly at all.
ii
Approved For Release 1999/Qpg : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
I. DEVELOPMENT OF SOVIET PHOTOG wMTRY
A. INTROWCTION
In any attempt to evaluate photogrammetrical mapping procedures in
the U.S.S.R. the fact that photogrammetry is a young subject which has
not yet reached maturity must be constantly kept in mind. Though photo
grammetrical surveys on the ground were made prior to World War I, aerial
photography - at first as an auxiliary aid to the mapping of detail by
conventional ground methods - became a major tool of the map maker much
later,
Only in the last twenty years have governments fully recognized that
mapping programs, to be efficiently completed, must utilize aerial photog-
raphy. Furthermore, it has only been in this period that photogrammetri-
cal theory, techniques and procedures have been developed to the point
where it has been possible to organize photogrammetrical survey operations
in a rational and economical manner suitable for mapping vast areas.
It would seem dangerous to assume from the fact that techniques and
procedures differ from country to country and vary in accuracy that this
is due to the technical superiority of one country over anoth4r.. Techni-
cal superiority may exist, but in the main the organization of photogram-
metry and the development of different techniques and procedures in dif-
ferent countries would seem largely to depend on their mapping needs and
whether these can be realized in a reasonably short time. These in turn
will depend to a great extent on the sizes of the countries and the den-
sities of their populations and, to a lesser extent, on their political
structure and industrial stability.
Approved For Release 1999/09/02 S '/,~. -,RDP79-01083A000100040001-4
Approved For Release
02 : CIA-RDP79-01083A000100040001-4
A good illustration of this variation in practice is the difference
between Western European and American photogrammetrical practice. Since
Western Europe conaists of a number of comparatively small countries, each
fairly adequately mapped before the advent of aerial photogrammetry,-the
effort there has been largely in the development of techniques, procedures
and instruments of greater and greater accuracy and for the utilization
of photogrammetry :or mapping on larger and larger scales. In the U.S.A.
the effort has beea devoted more to reaching an acceptable accuracy for
the mapping of large unmapped areas on comparatively smaller scales with
the emphasis on speed and economy of operation.
Moreover, the money appropriations for mapping comparable areas and
the number of properly trained and experienced photogrammetrists per unit
of area is higher in Europe than in America because of the over-all greater
density of population. This does not necessarily imply that the competency
of American photo grammetrists as -a whole is lower than in Europe, only
that it must be spread out thinner. Furthermore this justifies, and to
a certain extent enforces, the development of techniques and procedures
which can utilize comparatively unskilled labor in many of the photogram-
metrical operations.
Until quite recently, be-cause of the lack of a true appreciation of
these underlying causes, European photogrammetrists have considered their
technical competence as being superior to that of their counterparts in
the U.S.A. . But though the average European photogrammetrist may be
better trained than the average American photogramrnetrist in the theoreti-
cal aspects of the subject, themore broad-minded Europeans are now ready
to admit, from the practical standpoint of competency, that American
Approved For Release 1999OtO2: CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 StDP79-01083A000100040001-4
photogrammetry is on a par with that in Europe considering the different
mapping requirements in each region.
Overall mapping requirements and conditions in the U.S.S.R. are much
more comparable to those in the U.S.A. than to those in Western Europe.
However, the area of the U.S.S.R. - over eight and one-half million square
miles - is approximately twice that of the U.S.A. and the population, on
the other hand estimated at the present time to be less than 200 million,
is not very much larger than the population of the U.S.A. Thus, if there
is any validity in the foregoing argument one might expect to find (a) fewer
competent photogrammetrists per unit area in the U.S.S.R. than in the U.S.A.,
(b) that standard mapping scales are on the whole smaller, (c) that accuracy
specifications are less strict, and (d) that the techniques and procedures
and instruments that have been developed are cruder and more easily handled
by those not fully acquainted with photogrammetrical theory.
In many respect these conditions appear to have been true up until
recently, though again as in the case of the comparison between Western
Europe and the U.S.A. this does not in itself reflect on the over-all com-
petency of the photogrammetric profession in the U.S.S.R. Indeed one gets
the impression that photogrammetrical operations in Russia are extremely
well organized and that the facilities for training in the theory and
practice of photogrammetry are considerably better than in the U.S.A.
Furthermore there seems to be no doubt that that with the aid of photo-
grammetry the U.S.S.R. In recent years has had an enormously greater out-
put of topographical mapping than the U.S.A.
The reason for this state of affairs would seem to be fairly obvious.
The great period of industrial expansion in the U.S.A. took place before
Approved For Release 1999/09/02 :EQDP79-01083A000100040001-4
Approved For Release 1999/P : CIA-RDP79-01083A000100040001-4
the advent of- photogrammetry, whereas in the U.S.S.R. it is at its height.
Consequently in the U.S.S.R. there is an urgent and pressing need for maps
not only for military but for economic purposes. These needs cannot but
be heeded by the Soviet government. In the U.S.A.,, though the need for
the completion of a national map is fully recognized, its urgency is not
so apparent. Time, the budgetary allowance for civilian mapping in pro-
portion to other budgetary needs is probably far greater in the U.S.S.R.
than in the U.S.A.
In-spite-of this it can be reasonably inferred from the material re-
viewed that the supply of precision instruments and trained personnel has
not been sufficient for the demand in the U.S.S.R. up to the present.
Graphical methods seem to be still widely employed and the instructions
for mapping are apparently directed toward personnel not fully trained-or
highly experienced. Some of the mapping apparently is -still done by plane
table on the ground (Siberia). Much of the actual photogrammetry is ac-
compli-shed with the use of graphics approximating instruments. Also,
photographic materials and their processing do not seem to be as good in
the U.S.S.R. as in either Europe or the U.S.A.
The outstaniing exception to this situation would appear to be in
the advances the Soviets have made in the design and production of wide-
angle aerial camera lenses. These would appear to be of very high quality
in regard to the evenness of resolution and in the smallness of, their dis-
tortion. One has, of course, to treat the claims about these lenses with
considerable scepticism until samples are available for testing in this
country.
Approved For Release 1999 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 sgRDP79-01083A000100040001-4
B. HISTORICAL DEVELOPMENT OF
OTOGRAPIM TRY IN ' U. S .S .R
It is difficult to determine how far Soviet photogrammetry has de-
veloped independently from that of Western Europe and that of the U.S.A.
This is because of the propaganda line that always appears in any histori-
cal accounts of developments in the U.S.S.R.
This propaganda has noticeably increased since the end of World War
(1)
II. For instance, Veselovskiy in his book published in 1945 is quite
frank in stating that stereo-photogrammetric instruments of foreign make
have been used almost exclusively up to the time of writing. Skiridov,
however, in his book published in 19S1(2), though apparently well versed
in foreign methods and instruments, goes to great pains to claim priority
in development for almost everything connected with photogrammetry. In
his "short historical review of aerial scale photogrammetry in the U.S.S.R."
(page 6-11 of his book) there are some illuminating comments.
It is stated that in 1925 that the stereo-planograph was introduced
in the U.S.S.R. though at this point he does not mention the Zeiss Company
who were its originators. According to the writer its potentialities were
immediately recognized in the U.S.S.R., far more so in fact than in the
country of its origin.
1926 is given as the date when the first Soviet treatise on the an-
alysis of the exterior orientation of a single aerial photograph was pub-
lished. This was written by N. G. Ke11'(3) and Skiridov considers it to
be the original treatise on the subject. Nevertheless, this particular
and fundamental problem had been comprehensively analyzed long before,
notably by Finsterwalder in Germany in 1897, by Roussilhe in France in
1917 and by McCaw in Great Britain in 1922.
Approved For Release 1999/09/02 : JDP79-01083A000100040001-4
Approved For Release 1999/09RuriCIA-RDP79-01083AO00100040001-4
In 1928 S]ciridov himself published an analytical solution of the
problem, equally fundamental, of the relative orientation between photo-
graphs and the means thus provided for spatial photo-triangulation. How-
ever he was preceded in this, for according to 0. PonGruber, the theore-
tical foundation was laid by Finsterwalder as early as 1899 and practice
applications of the theorywere suggested by other German and French aci-
entists before the-first world war. Certainly a thoroughly practical
method was in eixistence before 1928, for instance in the work of H. G.
Fourcade publizhed in the Transactions of the-Royal Society of South
Africa in 1926.
1930 is the date given when Gapochko(4) started wor$ing out a method
of drawing relief on the aerial photographs themselves with the help of
stereoscopes and then reducing it to map form. An analogous method had
been completely worked out and put into practice in the U.S.A. in the
early twenties. This was the famous Broek method.
The leading photogrammetrical instrument designer in Russia seems
undoubtedly to have been and to be F. E. Drobyshev. He had, according to
Veselovskiy, designed a nine-lens camera in 1931. It would appear that
the most original photogrammetric instrument that the Russians have been
able to develop has been his stereometer. He started to work on this -in
1931 and the fi:rst instrument was completed in 1934(5, 6, 7). However,
it is worth noting that the instrument did not come into general use
until after Wor:Ld War II. From a mechanical point of view as noted later
in this report the principle features of the stereometer, of which there
have been several versions, is that both photographs in the instruments
are maintained in co planer relationship at all times-regardless of the
Approved For Release 1999 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : fi%JDP79-01083A000100040001-4
tilts which may have existed at the time of exposure; that the line of
sight of the optical viewing system is kept perpendicular to the plane of
the photographs at all times and that by means of ingenious mechanical
devices parallaxes are obtained from coordinate movements of the photo-
graphs in their common plane. No comparable instruments are produced in
this country or in Western Europe though in the last analysis the instru-
ments do not perform any task that some European and American instruments
do not also perform.
The claims that the Soviets were the first to produce satisfactory
wide-angle aerial camera lenses may be true. This development was due to
the work of M. M. Rusinov who produced in 1931(8) the "Liar objective"
with an angular field of view of 1000. In this connection it should be
noted that Skiridov states with pride that the later super-wide-angle
objective "Russar" with a field of view of 1220 has no counterparts out-
side Russia.
In 1936 under the leadership of M. D. Konshin and G. V. Romanovskiy
there were developed methods and procedures for mapping from aerial pho-
tographs on the scales of 1:50,000 and 1:100,000. Romanovskiy also intro-
duced at this time the "straight line" method for determining elevations
through a series of near-vertical photographs (see appendix VII). This is
undoubtedly an original technique. The principle feature in these mapping
systems is called the "differentiated method". It divides the mapping
process into two separate parts; first, the mapping of the planimetry and
second, the plotting of the relief. Great economy is claimed for this
procedure since it involves simplified instrumentation and lower standards
of training for those engaged in the work of applying it. Furthermore,
7
Approved For Release 1999/09/02 :SOIX RDP79-01083A000100040001-4
Approved For Release 19 2 : CIA-RDP79-01083A000100040001-4
the procedure can be stopped without wasted effort in those cases where
relief features are not required. On page 140 of this report the following
comment is made about this method "...(it) is similar to the Brock method.
It is the same in basic concept but has been extended and improved in some
important respects and in a few respects is not as good... It is ironical
that the only American mapping method (originating in America) - the Brock -
should have been neglected here but taken over by the Russians and that
one of the better F,uropean inventions - the Multiplex - should have been
neglected there bv.t taken over by us." Russia also claims to have devel-
oped further the Multiplex for purposes of aerial triangulation, using
extremely wide-angle field cameras.
In 1938 a method of spatial photo-triangulation based on computing
formulas by Zhukovwas adopted. This approach was one of the earliest to
be considered in id'estern Europe but was later abandoned with the intro-
duction-of the universal type of stereo-plotting instrument. It would
seem important then to note that this approach using stereo-comparators
and analytical comrputations has been given much consideration by the
Soviets in recent years while in Europe the emphasis stillis to a great
extent on the pureily instrumental or analog solutions. This is especially
interesting because there has been, still more recently, both in Europe
and the U.S.A. a re-examination and further development of the analytical
approach, induced by the introduction of electronic computing devices.
In commenting; about the most recent developments in photogrammetry
in the U.S.S.R., 5kiridov mentions the use of a multi-camera multiplex in
mapping mountainous regions* and also lays emphasis on the efforts being
First introduced for aerial surveying by Finland.
8
Approved For Release I 9 M2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 :sq~&-FDP79-01083A000100040001-4
made to determine the position of air stations and camera orientations at
the time of exposure by means of statescopes and mechanical pilots. Final-
ly it is briefly mentioned that position location by means of radar has
begun to be used. However, there is no elaboration of this statement in
the material under review.
II. SOVIET MAPPING SCALES AND PROCEDURES
According to Veselovskiy(1), in 1945 original surveys and map compi-
lations were being made from air photographs on six different scales,
namely: 1:10,000, 1:25,000, 1:50,000, 1:100,000, 1:200,000, 1:500,000.
The scale of 1:100,000 was the basic maping scale with 20-meter contour
intervals in flat country and 140-meters in mountainous regions. The
larger scales were used in highly developed areas and for special purpose
maps. The smaller scales were used for reconnaissance maps in undeveloped
and unexplored areas.
However, by 1951 according to Skiridov(2) , the decision had been made
to map the country on a basic scale of 1:25,000 with selected areas to be
filled in with scales varying from 1:2,000 to 1:10,000. This might be
taken to mean that the enormous task of completing the 1:100,000 scale
map of the country had already been effected. From other sources it does
appear that a great deal of mapping on the scale of 1:100,000 has already
been accomplished. It, however, seems inconceivable that mapping on this
scale is anything like completed, if one considers the area involved and
9
Approved For Release 1999/09/02 : LA DP79-01083A000100040001-4
E'r
Approved For Release 1999MM : CIA-RDP79-01083A000100040001-4
the fact that the mapping of the U.S.A. on a comparable scale (1 mile to
the inch) is not by any means finished. One is, therefore, inclined to
believe that except in highly developed areas, photogrammetrical proced-
ures will continue to be directed towards mapping on the scale of 1:100,000
for many years to come. Evidence to this effect is contained in the ex-
tremely comprehensive and detailed instructions contained in Ott I j of
the Instructions or Toi ograuhic Surveys on the Scale of 1:100.000(9).#
This manual which deals with photogrammetric compilations was first pub-
lished in 1942 and has been revised in 1950 to include many new procedures.
In 1945 Vese:Lovskiy also stated that the methods of compiling maps
on a scale of 1:100,000 were not regarded as firmly established. This is
certainly borne out in the second edition of the "Instructions" which is
replete with alternative procedures for undertaking identical operations.
Moreover, these reveal a tremendous effort to take care of all contingen-
cies and to-make 1.t possible to use personnel not fully trainedin photo-
grammetrical theory by relieving them of troublesome decisions when unex-
pected technical difficulties arise. In-several cases, for instance, the
Instructions insist that when a certain procedure does not realize the
required accuracy that the work must then be handed over for its solution
to the "Chief of the Brigade" or, in special cases, to the "Chief of the
Department".
A certain amount of freedom appears to be given to the local survey
supervisor on the methods to be employed in mapping any particular small
region. Guiding considerations in general are that in areas with differ-
ences of elevations of less than 300 meters per photograph the maps should
25X1A5a1
Approved For Release 1999 2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 :sGDP79-01083A000100040001-4
be compiled from photo maps or controlled mosaics on which have been drawn
contour lines obtained stereoscopically. In mountainous areas on the
other hand the contours are to.be drawn directly on the map from aerial
photographs by means of a stereoscopic plotting instrument, the photo-
graphs having already been adequately controlled.
Within this general framework the detailed alternative procedures
are numerous and it would seem that the guiding rule is to choose the more
approximate method, provided the necessary accuracy is obtainable, because
it is generally the simplest and requires less instrumentation. Further-
more, it is evident that the local survey organizations do not have a
free selection of instrumental equipment and must often make shift with
what is supplied them. This is still another indication of the lack of
standardization and of the shortage. of the most up-to-date equipment.
In order to obtain a definite picture of the great variety of pro-
cedures given in the "Instructions" the general subjects discussed therein
are summarized below. Detailed comments may be found in Appendix I.
The point is made in the introduction to the Instructions that the
newer alternative procedures can be introduced without disrupting the
technological program as a whole. Actually, the new procedures which were
introduced in the second edition include the methods which utilize stato-
scope recordings in spatial photo-triangulation, the use of a multiplex
for extending control, the use of the (RP-6) stereoscopic drawing appara-
tus and the transference by means of a single multiplex projector of de-
tail from the photograph to the map. Thus it will be seen that the inno-
vations are considerable and that there appears to be a trend towards
using methods which are more comparable to those in use in Europe and the
Approved For Release 1999/09/02 : a DP79-01083A000100040001-4
Approved For Release I 99Q$ 2 : CIA-RDP79-01083A000100040001-4
U.S.A., though the better features of the "differentiated method" are
maintained.
However, it would seem likely that those in command of photogramme-
trical operations and confronted with the policy decision of changing the
basic mapping scale from 1:100,000 to 1:25,000 must be faced with formidable
problems of- reorganization. Many of the more approximate and simple
methods will have to be replaced by thosewhich are more theoretically
correct and therefore more precise. They.must have on their hands, if
the accounts of the vast amount of mapping that has already been achieved
are correct, a great number of people well trained and experienced in the
simpler methods. These must be retrained. Such a reorganization of
method and retraining of personnel cannot be done overnight with efficiency
and without disrupting the general mapping program. It appears that the
Soviets are attacking these problems sensibly and introducing the new
techniques by degrees and as alternatives to the older procedures.
Approved For Release 1999I9Ii2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : 9DP79-01083A000100040001-4
APPENDIX I 25X1 A5a 1
on the Soviet book entitled,
Nastavleniye po topograficheskoy s"yemke v masshtabe 1:100,000, Chast' II,
Fotoerammetricheskiye raboty.
(Instructions for Topographic Surveys on a 1:100,000 Scale, Part II,
Photogrammetric Operations)
Second Edition, GUGK, 1950.
Note: Paragraphs and itemized numerations are identical with those in the
original text.
I. Introduction (No comment)
II. Compilation of Moto-Mg and Controlled Mosaics
A. Comuilatjon of the base horizontal control sheet
1. Where differences of elevation are more than 100 meters per
photograph.
2. For mountainous country in which case the nadir point or
its approximation must first be determined.
Various alternatives and procedures are discussed for undertak-
ing this part of the operation depending to a large extent on the availa-
bility of ground control points,. In general, the network of control is
built up by radial line intersections using principle point base lines or
nadir point base lines as conditions require. However, none of the alter-
natives uses the slotted template technique. At this stage special points
are chosen and their horizontal positions determined for the purpose of
photo-rectification which is the next step.
13
Approved For Release 1999/09/02 : 01DP79-01083A000100040001-4
Approved For Release
02 : CIA-RDP79-01083A000100040001-4
B. Rectification of Individual. Photoarauhs
Different instructions are given for this operation depending
on the type of photo-rectifier available. If the range in elevation per
photograph exceeds 100 meters a separate rectification is made for each
elevation zone or, when convenient, rectification is made to a slanted
plane which represents an average reference plane for the area. The method
of rectification in each case must be by the trial and error method using
the points which have been previously selected for rectification purposes.
C. !kU_AtipZ Photo- w h the Aid of %e Horizontal Control Sheet
D. Preparation of Controlled Mosaics
The distinction between a photo-map and a controlled mosaic is
not clear here because apparently in both case, the photographs must be
rectified.
III. Stereo-Photo&rammetric Methods for the Determination of Elevation
and the RepreEentation of Relief.
Six alternative procedures are given in detail in this section.
A. The Metbodd._of. -the -Central Scjent1fic Research Institute of
Geodesv. ~.erial-Surveying and Cartoraphv
This consists of the following nine, separate steps:
1. Preliminary planning, identification and selection of points
to be used, etc.
2. Determination of the-elements-of relative -orientation.
3. Determination of x parallaxes and air-base lengths.
4. Extension of the horizontal -control network.
5. Deterzination -of the corrections to be introduced into the
initial values of relative orientation.
Approved For Release 1 9/02: CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 slit RDP79-01083A000100040001-4
6. Determination of the elevations of the air-stations.
7, Determination of differences in the elevations of selected
points in the area.
8. Absolute orientation of the flight lines and the photogram-
metric determinations of the absolute altitudes of points
in the terrain.
9. Adjustment of the whole network including the elements of
orientation.
Relative orientation is done either on a stereometer or on a
stereocomparator; in the latter case more computation is involved. A
simplified method of extending control is allowable if the number of air-
bases between control points does not exceed six and provided the tilts
of the photographs of the photographs are well within a limit of three
degrees.
In the extension of horizontal control the nadir points are
computed whereas the pass points are obtained graphically.
B. This method is essentially the same as A except that reliance
is placed on a.preliminary analysis of statoscope recordings taken during
the flight and which enable the relative heights of the air-stations to
be determined.
C. In this alternative the wide-angle multiplex is used throughout.
When the stage is reached for absolute orientation if the closing errors
in elevation are small they are adjusted graphically. However, if the
elevation errors exceed 5 mm. on the scale of plot, computation is introduced.
D. This is the same as C except that statoscope recordings are
utilized. Apparently the multiplex wide-angle method is usually employed
15
Approved For Release 1999/09/02 : O1ADP79-01083A000100040001-4
Approved For Release 1999/0 : CIA-RDP79-01083A000100040001-4
for what are called "main flight lines". These appear to be those which
have at least three ground determined elevation control points and one
horizontal point at each,end.
E. This method introduces the "straight-line method" for determin-
ing elevations which apparently does not require preliminary relative
orientation of t;he photographs and enables vertical control to be extended
along strips and across overlapping strips. It is not used when the relief
exceeds 100 riete:rs per photograph.
F. This mi?thod uses the stereometer for extending the vertical con-
trol. As in E it is only used when relief does not exceed 100 meters per
photograph. In using the stereometer for this purpose, two different
procedures are described in detail.
IT, SketchipA Contour Lines with the Aid of Simple Stereoscopes EaviAR
Wide Fieldi3 of View
It is remarkable that fully automatic plotting instruments are not
used and that tho contour lines are sketched as a rule on the photographs.
When additional vertical control is needed it is obtained by means of
simple parallactte rules. Of course, as is noted in the Instructions,
controlled mosaics and-photo-maps which have been rectified by zonescan-
not be used in this procedure. Emphasis is based on the correct Interpre-
tation of the landscape by means of these interpolated contour lines es-
pecially from the, geomorphological point of view and the photographs
apparently are invariably also interpreted on the ground by-field survey
parties who utilize extra prints for this purpose.
A. Instructions are introduced here for the use-of Drobyshev's
16
Approved For Release 1999MM : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 3DP79-01083A000100040001-4
topographic stereometer. With this instrument the orientation of a pair
of photographs can be effected, provided there are at least four elevation
control points in the stereoscopic model.
V. Comx~ilation of the Base Mays
Several methods are described. One is a visual transference of the
material from the contact print to the photomap. Another is to transfer
the map data by means of a special stereoscope having variable magnifica-
tion so that photographic images and the photo plans may be seen stereo-
scopically at the same time. The contour lines are then drawn in by hand
on the photo plan. In mountainous country a single multiplex projector
is used for transferring contour lines and other details which have been
drawn on the photographs to the base map. This involves raising or lower-
ing the projector as the work proceeds in order to adjust for scale varia-
tion. Finally if the multiplex projector is not available and a photo-
rectifier projector has to be used scale adjustments are computed by
special formulas.
VI. Co lation of a topographic ma with the help of a s ereoscc is
drawing instruteA .
This is the one and only case in the Instructions where the contour
lines and planimetric details are drawn directly from the photographs onto
the map. The instrument, (RP-6), is quite simple and is virtually a
stereoscopic camera lucida. The observer sees the plastic image of the
landscape and at the same time the point of the plotting pencil.
Approved For Release 1999/09/02 : OIAJRDP79-01083A000100040001-4
Approved For Release 1999M ?M2 : CIA-RDP79-01083A000100040001-4
APPENDIX II.
Review of Potograummetriya (Photogrammetry), by N. N. Veselovskiy(i)*
Published in Moscow in 1945 by thee` Geodetic and Cartographic Literature
Publishing House at the Co 2 X1A5aeples Commissars of the U.S.S.R.
Note: The reviewer has no-familiarity with the Russian language. He has,
however, in addition to a photostat copy of the original text been
provided with an unedited translation of the very detailed table
of contents and selected text passages of seeming importance,
1) Scope Book
The aim of the book is to give a rather full account of photo-
gramrnetric theory and contemporary Russian practice. In the author's
.introductory note however, he states that certain problems have not been
dealt with fully or at all. These include methods for determining exterior
orientation Burin,; flight, and film distortion and its influence on the
precision of photogranmetrical processes.
The book was published at the end of the war with Germany and. was
based on a series of lectures which were probably given in wartime. As
such, it is a pra:Lse-worthy effort, as good if not better in its organi-
zation and co:l:pleteness as the first edition of the Manual of Photogram-
metry, published by the American Society of Photogrammetry at about the
same tir.e (1944). The book is perhaps more comparable to the Professional
Papers of the British Air Survey Committee which were published over a
period of about twelve years starting in 1925. In this reviewer's opinion,
the presentation cf the geometry and mathematics of photogrammetry is
better balanced and more complete than that in any of the contemporary
25X1A5a1
18
Approved For Release I 99 2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : 9 -,DP79-01083A000100040001-4
textbooks on photogrammetry written in English for student consumption
and this includes those by Hotine and Hart in Great Britain and Talley
and Bagley in the U.S.A. Much more attention is paid to the problems of
error accumulation. Nevertheless, the methods for adjusting aerial tri-
angulation would seem to be not fully developed theoretically and compar-
able to those suggested by Hotine In Professional Paper of the Air Survey
Committee, No. 7. (British)
There is, of course, evidence of national bias in the book. Much of
the photogrammetric development, described as being original in the U.S.S.R.,
appears to have been copied in whole or in part from the work accomplished
in other countries.
The book might have been more carefully integrated. Ground photo-
grammetry is discussed at the end and one gets the impression that this
section was added almost as an afterthought. The spatial extension of
control is treated quite independently and much later in the book than
the methods of extending control horizontally by radial line methods -
there being interposed the whole subject of photo-rectification and the
making of photo maps. However, this seeming divorce of the two principal
techniques for extending control is perhaps understandable when the general
system of producing maps on various scales in Russia is appreciated. This
is described towards the end of the book (Chapter 31).
2) Methods of Mapping on various Scalesin the U.S.S.R.
Original surveys and map compilations were being made in 1945 on
the following scales: 1:10,000, 1:25,000, 1:50,000, 1:100,000, 1:200,000
and 1:500,000. For scales of 1:50,000 and larger, hypsometry was obtained
19
Approved For Release 1999/09/02 : g4DP79-01083A000100040001-4
Approved For Release 1999109102 : CIA-RDP79-01083A000100040001-4
almost entirely b:r ground field parties. In areas of no great relief
these-parties wero supplied with photo-mosaics and the process of deter-
mining elevations carried along at the same time as-field checking and
geographical interpretation. Apart-from supplying detail the aerial
photograph was used for supplementing ground control by means of horizon-
tal radial line traverse. In mountainous country the maps were usually
constructed from the individual photographs without the intervening mosaic.
It is, however, acted that at the time of writing successful experiments
had been undertaken in drawing relief features by means of the stereometers
of Drobyshev.
On the scale of 1:100,000, photographs from multi-lens cameras were
previously used, but because of the complicated procedure they were dis-
carded in favor of single-lens photography taken with wide-angle objectives
having focal lengths of 100 mm.
It is stated that methods of compiling maps on a scale of 1:100,000
cannot be regarded as firmly established and a varietyof methods are
described,-some suitable for flat country and sortie for mountainous country.
In all methods great reliance is placed on ground field work- consisting
mainly of tachyometrical-traverses for tying in local radial line aerial
triantul.ations. In flat country these traverses are run perpendicularly
to the flight lints and spaced about four stereoscopic pairs apart. In
mountainous country, spatial photo-triangulation is often used for eleva-
tion control, but this procedure apparently does not altogether eliminate
the usual horizontal radial line traverses and their control by ground
traverse or plane table survey. Discontinuous form lines are often sketched
on-the photographs with the aid of simple stereoscopes and adjusted later
to the proper contour interval in the field.
Approved For Release I 9/ 02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
Methods for the compilation of maps on 1:200,000 scale from aerial
photographs are less developed than for the larger scales. It is concluded
that the basic method for compiling planimetry should consist in assembling
uncontrolled mosaics from single-lens photography, the scale being obtained,
apparently, largely from navigational data. (This statement is the review-
er's own interpretation of what in the translation appears to be a rather
involved procedure.) Methods for obtaining relief appear to be similar
to those employed for the.1:100,000 scale. The need for geographic inter-
pretation and generalization is stressed.
No methods of utilizing aerial photographs for the 1:500,000 scale
maps had been worked out in 1945, and the author states this as extremely
unfortunate because'this scale is a basic one for sparsely populated re-
gions In the U.S.S.R. However, a scheme depending on the construction of
a special four lens camera is given, which would appear to be basically
sounder than the trimetrogon system. All four objectives have their optical
axes parallel; one is for vertical photography, two, lateral obliques and
one for. forward and rearward pointing obliques. The field of view is de-
flected onto a common film base by means of prisms In the case of the
oblique photograph. In order to bring the scales of the obliques more
nearly to that of the vertical photograph the focal length for the vertical
picture is 100 mm, while that for the obliques is 160 mm. The photography
in the direction of the line of flight is used to increase the precision
in relative orientation and longitudinal tilt (x tilt).
Aircraft is used`at speeds up to 400 km an hour for time-saving pur-
poses and flight lines are up to 100 km in length. As a general practice,
spatial triangulation is used for obtaining elevations, bridging being
Approved For Release 1999/09/02 ScIA RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
contemplated an to 22 km when elevation tolerances of 20 meters are accept-
able, and up to 90 tin if the tolerances are 40 meters. A parallactic grid
stereoscope is used for relief interpretation. The approximate and so-
called "straight-line method" of control extension is used in flat regions
and a specially designed rectifying printer is supplied to aid in the
transfer of planimetric detail to the map.
It is of importance and great interest to discover whether this type
of reconnaissance mapping has by now been put into routine practice or
has been abandoned because of unforeseen difficulties.
To this reviewer the matter of most interest in this summary of
Russian mapping activities is that the Russian cartographers in 1945, while
realizing the need. for utilizing aerial photography to a greater extent
than previously, have nevertheless developed methods for larger scales
which are dependerLt.on a combination of photogrammetry, ground field check-
ing and detailed ground surveying. This is perhaps understandable because
a map is lifeless if it does not reflect a close-up knowledge of the ter-
rain and the humaal activity thereon. In America this can largely be gleaned
second-hand from a more or less literate population, whereas in the U.S.S.R.
such geographical intimacy must be obtained first-hand. So it may be sen-
sible to organize the mapping activities in the ways described. Further-
more, if this interpretation of the situation is correct then it partially
explains why the only really new photogrammetrical instrument developed
up to 1945 in the U.S.S.R. is Drobyshev's stereometer. This measures rec-
tified x and y parallaxes and can be used in a comparatively simple manner
for orienting pairs of photographs provided the elevations of at least
five points imag,sd in the overlap are known.
Approved For Release 19 2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 siO14 RDP79-01083A000100040001-4
In connection with the general viewpoint the following broad state-
ments are made (referring to the orignal text).
p. 14, "Development of stereo-photogrammetric operations in the U.S.S.R.
has been slow due to lack of instruments chiefly those necessary to pro-
cessing the photographs. While instruments of foreign make have previous-
ly been used, recently as a result of much work, experimental stages have
been passed through and instruments for mapping on scales of 1:50,000 and
1:100,000 are being produced in the U.S.S.R. Methods combining ground
survey and aerial photography have been more developed than any other and
have been widely used for mapping on scales of 1:10,000 and 1:25,000."
p. 91, "Determination of exterior orientation by photogrammetry is one of
the basic problems at the present moment.'
P. 173, (After an analysis). 'Thus we see that the precisions of analyti-
cal and graphical radial line triangulation are close to one another but
the former requires much more photogrammetric processing and computation."
p. 278, "Determination of elements of exterior orientation in flight is
of greatest importance in the development of contemporary photogremmetry.
Research work in connection with the application of gyroscopic instruments
is in progress. If the vertical position of the optical axis can be es-
tablished with a precision of from ten to twenty minutes, rectification
of aerial photographs will become unnecessary and will be replaced by
bringing them to a predetermined scale."
p. 296, "Spatial photo-triangulation becomes simpler when it is used in
Approved For Release 1999/09/02 : 08DP79-01083A000100040001-4
Approved For Release 19990 : CIA-RDP79-01083A000100040001-4
conjunction with plane radial triangulation and when the elevations are
determined by ordinary (?) methods."
P. 309, "Instruments-of the universal type such as the Stereoplanigraph
of Zeiss are not popular in the U.S.S.R., and their number is greatly
limited because of their low productivity in extensive cartographic work.
The use of this type of instrument is restricted to the compilation of
large scale special purpose maps and for spatial photo-triangulation. By
splitting the photogrammetric process up into steps a larger productiveness
is obtained than by the use of the universal methods, and the process
becomes more economical, It is therefore widely used in the U.S.S.R. for
planimetric mapping on intermediate and small scales."
p. 388, "Methods :for the revision of old maps by aerial photography are
not sufficiently developed. There can be no doubt however, that this prob-
lem is about to bocome very real and equal in importance to that of map-
ping unsurveyed areas."
3)
Mis cellaiLeous Information} and Comment
a) Lenseis and Aerial Cameras
It is stated that the creation of a wide-angle objective
with sufficient light intensity and no appreciable distortion was consid-
ered by German and. American experts to be quite impossible. Nevertheless,
the Leningrad Institute of Aerial Photography started making the attempt
and produced the first sample in 1934. (Liar-6). This had considerable
optical drawbacks - though its resolving power was claimed to be satis-
factory - it had distortion, chromatic and spherical abberations, astigmatism,
21f
Approved For Release 19ff O2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 sRDP79-01083A000100040001-4
and bad vignetting effects. (!) The Russar type of lens was first produced
in 1936 but its resolving power was low. (9-12 lines a mm.) However, the
new types of Russar lenses are considered satisfactory for small-scale
aerial stereoscopic photography. Note that there is no claim that these
lenses are satisfactory for large scale mapping. A comparison was made
of Russar objectives with the Zeiss Topogon. Results given in the table
below show the disadvantages of the latter.
Resolving
Power in
lines to 1
Size of
mm. at the
Relative
Angle of Photograph
edge of
ecti;
Aperture
View in ? in c
image
Distortion
in 0.01 mm
at the edge
of image
H
l
100
1:5
3
140
18 x 18
5-7
a
ussar -
.
Russar-5
120
1:4.5
104
23 x 23
15
5-7
Russar-19
100
1:5.3
104
18 x 18
15
- 20
5-7
Russar-22
70
1:8
122
18 x 18
15
- 20
1-2
Tafar
1:4.5
70
18 x 18
40
- 60
Less than 1
Topogon
1:6.3
93
18 x 18
20
- 25
25 30
In this table the resolving power and distortion of the lenses are given
only at the edges of the image. This does not constitute a fiar compari-
son. The earlier cameras built by the Russians were plagued with shutter
trouble. Of historical interest is the statement that a 9-lens camera
designed by Drobyshev was constructed in 1931.
(b) Aircraft for Aerial Photography and Navigation
A table is given in which are shown the basic characteris-
tics of aircraft available for aerial photography in 1945. Of the six
Approved For Release 1999/09/02 :tDP79-01083A000100040001-4
Approved For Release 1999M : CIA-RDP79-01083A000100040001-4
types mentioned, only one - the P-Z Sesquiplane - had -a practical ceiling
of over 16,000-feet. Considerable attention is given to navigation for
aerial photography, and apparently course indicators of the sun compass
type and statosco:pes of the Mendeleyev type (Finnish?) were in use at the
time of writing. The preparation of flight charts from the best available
mapsof the region being flown is stressed and it is strongly recommended
that such maps be given additional emphasis in bright colors, for features
-such as forests and highways, to facilitate the general orientation for
the pilot.
(c) Original Instrumentation and Techniques
It is evident that up-to 1945 the Russians had not succeeded
in producing many instruments which were not more or less copies of foreign
instruments. A notable exception is the stereometer of Drobyshev previous-
ly mentioned.
Much attention is paid to photorectification and very little to the
universal type of stereoscopic instrument. It is stated that the Zeiss
multiplex is sometimes used for extending elevation control and that the
Canadian grid method has its application in certain parts.. of Russia. The
Brock method of mapping is briefly described and one surmises that it has
influenced Russian thinking to a considerable extent.
The Stereo-Universal of Skiridov for obtaining relative orientation
is, claimed as an original Instrument though it is admitted that it is
similar-in -concept to Fourcade's stereogoniometer. The topographic stereo-
scope of Romanovsk:Ly is very suggestive of the Barr and Stroud topographi-
cal stereoscope. '.['he double projector of Drobyshev seems to be a descend-
ent of Fourcade's double projector described in 1940 in the transactions
26
Approved For Release 199510=2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 :SqM- DP79-01083A000100040001-4
of the Royal Society of South Africa. Graphics seem well advanced and the
methods of radial line triangulation described probably have been influ-
enced by the Finnish work on this subject.
27
Approved For Release 1999/09/02 : CpP79-01083A000100040001-4.
Approved For Release I 99q f,2 : CIA-RDP79-01083A000100040001-4
APPENDIX III
25X1A5a1
General Discussion
of
11Dde , bm Z, , Chapter 4
"Saxll Scale Aerial Photography" (1Q)
by Konchin, M.D., Rusinov, M.M,, Yutsevich, Yu.K. and Sokolova, N.A.
A. Summary and Conclusions
This is a medium-detailed handbook or encyclopedia on mapping methods
and instruments. The theory appears to be well-developed, and current with
ours in the main, except possibly in-the error theory. Accumulation of
error is mentioned and formulas are given, but they are simple, and I doubt
that careful theoretical error studies have been made. The importanceof
error accumulation has only recently been realized here, however, so if
the U.S.S.R. is bohindin this respect it is by only a year or two. In
the experimental Lnvestigation of errors and accuracy of primary measure-
ments theyseem to be a little ahead, probable errors for-the various set-
tings and material distortions being given. For example, it is stated that
the error in para:Llax measurement caused by image motion has been found
experimentally tobe 120 of the computed amount of motion. No such figure
has-been measured in the U.S.A. I cannot tell, of course, how reliable
their-results are, but at least. the importance of such quantities is
recognized.
In general the mapping methods used are cruder than ours, using more
graphics-and-computation andfewer advanced instruments. This appears to
25X A5a1
28
pprove or a ease 199N@4@2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
be due to two things: the materials are not as good as ours, and the gen-
eral run of available instruments is of a lower order of accuracy and pro-
ductivity. The latter may be due to shortage of good design engineering
or production techniques; the reason is not known, but the statement is
made once that a certain instrument was not in production due to "techni-
cal difficulties*.,Statements are made, of course, that the Russian instru-
ments are simpler and therefore better, but this may be discounted. At
any event, most of the instruments are simple -- for example, telescope
optics are not used much in the stereoscope.
The result is, that the graphical methods and use of computation ap-
pear to be advanced further than ours. The methods they use for 1:50,000
and 1:100,000 maps are similar in structure to the methods we use (in tri-
metrogon work) for 1:500,000 and 1:1,000.,000. The organization of the work
is very reminiscent of our Brock method, and is good. From the descrip-
tions given, however, the accuracy obtained in 1:50,000 mapping must be
less than we have in such maps. And the amount of field control and office
work is greater than we need. They seem to fly somewhat lower than. we do.
The cameras used are wider angle than ours, and on the surface the
lenses appear to be better. This is fictitious, however. The lenses are
rated f/5.6 and f/6.3 but I doubt that they work at better than f/8 or
f/li. This is implied by the shutter speeds used: 1/50 to 1/100. A small
part of the slow shutter speeds may be caused by slow film (probably slower
than ours, since the rated resolution is 60 1/mm) but most of it must be
required for small lens stop. Hence the lenses are most likely f/8 or
f/11 lenses, with a large enough opening so that they can be rated f/5.6,
but will not work there. This is conjecture. Complete test data on reso-
lution and distortion would be required for a reliable evaluation.
Approved For Release 1999/09/02 : &iA-RDP79-01083A000100040001-4
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
One instrument they have would be useful to us: a-small rectifying
projector using-reduced diapositives, for use where we use the vertical
sketchmaeter. The idea is not new to us, but we have never gotten around
to making one.
In summary, their mapping methods are ingenious and theoretically
sound, cruder thF.n ours, almost certainly less accurate and less -efficient,
with a shortage of good laboratory instrumentation, and a large use of
simple mirror stereoscopes rigged up to do a little more than our simplest
instruments. There is some disagreement between different parts of the
chapter as to whether certain lenses are or are not in production.
B. Precis and D}.scussion
I,, Small Scale Aerial Photography
a. Introduction,
Need for aerial mapping to replace ground methods: cheaper, especially
in wild areas; possibility of reconnaissance mapping over large areas;
use of less highly trained people. This reads very much like our discus-
sions on the samo subject.
b. Compilation of maps at 1:50,000 and 1:100,000 scale.
1. Aerial Plhotogr~phy. Wide fields desirable for precision and ef-
ficiency, but multi-lens cameras complicate processing and not precise
enough. Hence, ride angle lenses used: Russar-l, Russar-22, Topogon.
Latter discarded for having too much distortion. Russar-l: 100 mm F, 1000
field. Russar-22: 70 mm F, 1200 field.
Discusses effect of material and equipment distortions on parallax
and height measurement. Of note: 30
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
Approved For Release 1999/09/02 : Cf P79-01083A000100040001-4
Photo-papers shrink 0.3%, a little worse than ours.
Differential film shrinkage is 0.1%, compared with 0.03%
for ours
Accidental local film distortion averages .04 mm, compared
with about .01 mm for ours.
Glass diapositives, or photo-paper pasted on glass is used
for more precise work (1:50,000).
Emulsion:! resolving power 60 1/mm, aerial film.
Resolution of Russar-1: 12-15 1/mm. Russar-22 not given.
Anti-vignetting filters are used.
States that all errors are proportional to size of photo
or camera focal length; therefore using larger cameras
would give no improvement. This is false.
Frame size 180x180 mm. Russar-19 and -la used for 1:50,000
maps, Russar-22 for 1:100,000 maps. (But note statement
later, that Ru.ssars above-16 have not been produced).
Photo scale 1:30,000 for 1:50,000 maps
Photo scale 1:50,000 for 1:100,000 maps. Note that this
means altitude of 3500 meters.
Exterior orientation records kept:
Statoscope (?2 in. accuracy)
Horizon, through auxiliary camera chamber attached to
camera. Sometimes two. Accuracy t 7 to 10 min.
Level bubble -- auxiliary indications.
'Watch -- gives rir base to +3% .
Sun photographs, for angular orientation. Ref. made
to Santoni cameras, no Russian make mentioned.
Experiments in progress on radio altimetry (interference)
and gyroscope tilt determination.
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
2. of raametric Control Methods.
*Straighi;-Line Method* for-low relief.
Spatial 9'.'riangulation for high relief.
The straight line method is an ingenious method for bridging ele-
vations, using two adjacent and overlapping strips. It computes elevation
differenceu by the standard parallax formula, assuming the photographs are
level,-and requires an (uncontrolled) mosaic of the strips. Hence it has
only limited accuracy, theoretically. Work Is graphical and computation.
Stated to be used. where relief does not exceed 500 to 100 meters per model,
and can bridge 4 to 5 models for 1:100,000 maps, no bridge at 1:50,000-
-It can also be used to interpolate elevations on a single model.
Spatial i hototriangulation. Used widely for 1:100,000 maps. For
1:50,000 maps only in mountainous areas because distortions cause too much
error for flat country. In flat country, "Method of Continuation on Stere-
.ometer* is used. Steps in Spatial Triangulation are as follows:
(1) Measure y-parallaxes in stereometer.
(2) Comp to relative orientation, and convert to pseudo-absolute
with respect to an arbitrary plane.
(3) Measure x-parallaxes and compute elevations. Adjust to
control.
(4) Meanwhile, triangulate planimetric positions using nadir
point In mountains, or nadir or principal point in plains.
Probably a radial line plot, though it doesn't say so. This
gives air bases for step (3).
Also, the "Lateral Traverses Method" is widely used for 1:100,000 scale.
This is apparently a mosaic of a single strip, holding to the center points
for direction.
Also, the Multiplex is used for bridging. From the tenor here
and in other places it seems that there are few multiplexes.
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SE RE
Approved For Release 1999/09/02 : CIA-TRDP79-01083A000100040001-4
3. Planimetric and Topographic Mapping
(1) Planimetry. Shows that pictures are not maps, due to tilt,
scale and relief displacement. For plaimetric maps of flat country one
could rectify and ratio to horizontal control, bu tbis _U too complex
and is not used. d. Similarly for planimetric maps in mountains one could
rectify to an inclined plane, or even to several such planes in zones on
one ploto. In general, to correct for relief displacement one must vary
the enlargement at each elevation. The portions could be assembled into
a mosaic, but this gives too many pictures. Hence, it is easier to vary
the ratio in a projector, and draw from the projected image.
This is not usually done on the rectifier, rather done with
a projector and pantograph. (Is there a shortage of rectifiers?)
This sounds like the system of laying a radial line plot, pro-
jecting and tracing, but dressed up with theory.
Also they have an instrument similar to Vertical Sketchmaster,
but the perspective relations of the photo not maintained. Statement that
parallax is eliminated is not true. Instrument is more complicated than
sketchmaster, probably easier on the operator, but no better in'function.
Enlargement of photo is not possible.
Small rectifying projector, using reduced diapositive, also
used for sketching operation, in conjunction with pantograph. T is we
shy d have. Up to 4x reduction.
(2) Topography at 1:50,000. For topography, Topographic Stere?-
eter is used. This is a stereometer like the Brock (but with simple optics
mirrors only) but fitted with mechanical corrections to reticles and par-
allax displacement, to correct for tilts. Photos remain flat. Uses contacts
33
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/01 :CIA-RDP79-01083A000100040001-4
paper prints pasted on glass. For maps of flat areas at 1:50,000,
5 elevations necessary in each model. (Note: Fig. shows 7 points.) Orient
to known elevations, then-contour on the photos.
required
Large amount of -field control/lowers the efficiency of this.
Hence, for 1:50,000 maps, extra operations are undertaken. They measure
y-parallax and compute relative orientation; also, use "Straight Line
Method" or othere, to compute elevations of center points. Hence, field
-control necessary only along sides of strip. But this means C-factor must
be low. We found. in Brock that this doesn't work -_ and the Brock equip-
ment is much more precise. Probably no more than 500 C, and probably more
nearly 300 C.
Also describes scheme for using two strips jointly with control
along side lap of every second strip. Properly adapted this might work
in some of our methods.
Note: suggests mean error of parallax difference measurement
.03 mm. I do not believe this possible with instrument described. Impli-
cation is made that the method works at 600 C, if bridging is not required.
The contours are drawn on the photos and corrected for tilt
and relief along with the planimetry. This is similar to some parts of
the Brock method.
(3) Relief Maps at 1:100,000. Relief is based on methods of
Spatial Triangulation or Straight Line Method, giving a net of elevation
control, at least 6 on each model, plus any "characteristic" relief points.
Then contours are drawn on the prints, by interpo2.ation between elevation
control points -- really, form lines -- not by Stereometer.
34
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
Can later be transferred to a mosaic, or a manuscript. Describes
stereoscopes:
C.vclops, half a stereoscope at a 300 slant
Chamber Stereoscope: fixed base with movable mirror stereoscope
and adjustable for scale differences up to 20
Barr-Stroud topographic stereoscope less often used. Corrects
tilts by tilting pictures on ball and socket joint, like our
KEK and has reticles above picture plane, like KEK, but they use
threads, and reticles are separated horizontally for parallax
changes.
Stereoscopic Drawing Instrument. Stereoscope with half-silvered
central mirrors, like Spurr's gadget. Photograph tables have tilt and
swing. Use a "multiplex table" for drawing. Draws in orthogonal projec-
tion.
(4) Geodetic Work. Which means ground control, or field work,
Low order triangulation used in mountains, and "geometric nets" (?) in
plains and open areas. "Tacheometrical" survey in closed areas. Elevations
are run by alidade traverses, and run up to 25 km. between control, or 15
km. from control to a pass point.
Also barometric leveling. Use one-base method, out to 10 to
20 km. for 1:50,000, more for 1:100,000.
In mountains, sometimes measure vertical angles from a control
point, and pick horizontal distances from the radial line plot.
Sometimes the surveyor takes the photos and a stereoscope into
the field, and draws the contour lines on the photos while running his
levels. Also, sometimes the planimetry is done from aerial photos and
the contours by plane-table.
35
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
c. Small Scale Cartography
By compilation if material exists. If not,-reconnaissance photogra-
phy, using multi-]. ns cameras: 4-lens TAFAM (USSR), or Fairchild T-3-A.
TAFAM: 1 vertical, 100 mm. F,84? lateral field; 3 obliques,
700 tilt, 2 lateral and 1 backward, 170 mm F gives 56? lateral
field. States 195? lateral field; hence almost no overlap in
chambers. Photographs the horizon. Fly at 4 kn., strips 30 km
apart.
Using TAFAM, :map compiled by graphic-computation methods described
earlier, since rectification of 700 tilts too hard. If T-3-A is used,
pictures are rectified.
Omitted pages 228-245 apparently describe camera perspective, light
loss, vignetting, lens characteristics, calibrated focal length, lens
aberrations, and several lenses.
(translation is sketchy from here on)
II. Wide Angle Objectives
Russar-1 apparently has small distortion (4.05 mm to 500) but very
large curvature of field at f/ 10, even though lens is rated f/5-7. Russar-
16 is very wide angle (1260) and looks good except (a) very large field
curvature and only f/12. As of this date (1949) the later lenses, Russar
19-26, have not r~aached production.
III. Aerial Cameras
MAFA Aerial Camera: Russar-1 or Hussar-19, 98 mm F, f/6.3, 1100
field only 104? used. Shutter speeds 1/60 to 1/180 or 1/45. to 1/130, 0.9
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
Approved For Release 1999/09/02 : C,P79-01083A000100040001-4
efficient, frame size 180 mm sq., using 190 mm film in rolls 22.5 meters,
giving 150 exposures. Intervalometer works 10 to 120 sec. Gross wt. 80
Kg. Camera uses both vacuum, pressure and edge clamping to flatten film.
Apparently camera has metering trouble. Also, image plane with fi-
ducial marks is on magazine, not cone, and registration trouble arises.
Remark: "The construction of aerial cameras is quite simple." Later
models moved the film plane to the cone.
Topo Camera Tafa- : Two cones:
Russar-19, 98 mm F, f/6.3., 11Q? field only 1040 used shutter speeds
1/25 - 1/75, 90% eff.
Tafar, 200 mm F, f/4-5, 63? field, shutter speeds'1/50 to 1/150, 80%
Image frame 180 mm sq., plus 15 mm for statoscope, level, watch and
data card. Roll size 190 mm x 50 m, gives 250 photos. Flattening by
suction and edge clamping. Intervalometer 5 to 110 sec. Weight about 40
Kg.
Item: when film edge is clamped, needles pierce the edge of the film,
so that you can tell where to cut the film in the darkroom, if not used up.
This was the first Soviet topo camera, and has proved satisfactory.
Topo Camera Tafa-2. Russar-19, 127 mm. F, f/6.3, 230 mm. sq. image,
240 mm x 50 m, film, 200 plotos. Shutter speeds 1/25 to 1/100, 90% eff.
Descriptions of mapping instruments, fragmentary translation.
Statement: Multiplex is used in USSR only for establishing control
in bridging.
Approved For Release 1999/09/023.t(G1-RDP79-01083A000100040001-4
Approved For Release 1999/O NUT CIA-RDP79-01083A000100040001-4
IV. Photogrammetric Instruments
r, )h of Drob shev. Mechanical restitution. Photos stay
Zftq&rzij,
horizontal. None manufactured as of 1949, due to technical difficulties.
38
Approved For Release 1999/09/Qci11A-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
SECRET
Discussion of Stereofotogrammetriva, (8tereo-Photggramiaetry) by
A. S. 51AMa1 ), 195
A. GEIFERAL S TA'IT1NT
This book was not fully translated, which made evaluation difficult.
However, the chapter and section headings were translated, the book is
profuse with figures and formulas, and some knowledge of methods, instru-
ments and terminology had already been gained from reading Geodeziya(.Ap-
pendix III), Drobyshev (Appendix V) in translation. Hence, I feel that
I was able to get a fairly good understanding of the book.
B. GENERAL REVIEW OP METHODS
The most important general methods of mapping used in USSR seem to
be: plane-table, differentiated methods, and universal instrument methods.
All of these appear to have major usage. The differentiated method seems
to be used more than the universal, and to be the pride of the Russians.
The multiplex does not seem to be used at all as a mapping method, but
only for spatial triangulation.
The plane table we are not interested in. The universal method is
about the same as we know It. The method of differentiated processes is
similar to the Brock method. It is the same in basic concept, but has
been extended and improved in some important respects, and in a few re-
spects is not as good. Thus we have the following tabular corkcarison of
25X1A5a1
SECRET
Approved For Release 1999/09/02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/?9 2r: CIA-RDP79-01083A000100040001-4
the major methods of mapping by aerial photography:
Method
U.S.
Europe
USSR
Universal
Multi
l
some
major
considerable
p
ex
B
k
major
some
negligible
roc
l
negligible
none
major
ow-order
some
sons
some
(-contour-finder)
It is ironical that the only American mapping method -- the Brock --
should have been neglected here, but taken over by the'Russians, and that
one of the better European inventions, the Multiplex, should have been
neglected there 'but taken over by us.
The Photogra=etric Theory is well developed in the U.S.S.R. The
theory is as well developed here, but the major difference seems to be
that there is application of the more advanced theory here, and the
knowledge is not wide spread. In the U.S.S.R. the more advanced theory
is contained in this textbook, even to calculation of higher-order effects
and how to apply them in orientation of photos, and so apparently-taught
in the higher courses.
The major trends in Russian developments seem to be:
1. A recognition of the importance of the optics, and a
development along those lines far exceeding our own,
if the reports on their lenses are true.
2. The tendtancy to make the corrections for orientation
mechanically, rather than byhomolog or optical means.
3. The use of numerical computations where we tend to
compute by analog, or not to compute explicitly at all.
A major crit:t:.cism is a lack of recognition of the importance of re-
dundance, or the use of redundance to simplify the computations rather
than to reduce the errors. This tendency is noticeable in American work
as well.
4o
Approved For Release 19999
9102 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : 4iDP79-01083A000100040001-4
RUSSIAN MAPPING PROGRAM
Until about 1951 or 1952 the basic Russian nkap scale has been 1:100,000
with 20 m. contours in flat areas, 40 m. contours in mountainous regions.
In the sparsely settled, outlying areas the publication scale is 1:100,000
and contours are 40 m., but the planimetric accuracy requirements are re-
duced to be equivalent to 1:200,000 scale. Apparently, from remarks made
later in the book and from remarks in Geodezi3a, there must have been
auxiliary mapping at 1:50,000, and some large scale mapping.
Starting as of now the future plan is to map the country at 1:25,000,
with selected areas filled in at scales between 1:10,000 and 1:2,000.
D. REMARKS ON INSTRUMENTS
Instruments remain pretty much as discussed elsewhere (Geodeziya and
Drobyshev) with one exception. The 70 mm. Russar 22 aerial camera lens
appears to have come into use by this time. The ultra-wide angle multi-
plex to use the 70 mm. pictures in triangulation appears to be available
and used. The Stereometers do not appear to have been adapted for 70 mm.
yet: they are constructed for 100 mm. photos, with a minimum setting of
90 mm. Skiridov gives correction formulas for using the existing stere-
ometers with 70 mm. photos. Also, the photo scales for the smaller scale
maps have been increased. Geodeziya gives photo scale of 1:50,000 for
maps at 1:1000000, and 1:30,000 for maps at 1:50,000. Skiridov given
ranges 1:55,000 - 1:75,000 and 1:30,000 - 1:40,000 respectively, In each
case the upper end of Skiridov's ranges would correspond to flying at the
same height but using a 70 mm. lens instead of 100 mm.
41
Approved For Release 1999/09/02 sj j rRDP79-01083A000100040001-4
Approved For Release 199WON02 : CIA-RDP79-01083A000100040001-4
Also, Skiriclov describes as a separate type of instrument, which we
will call Generalized Instruments, certain particularly Russian instru-
ments. These employ the Russian mechanical corrections for orientation,
but perform the junctions of the Universal type instruments. They seem
to be either the better stereometers with pantograph arrangements for
changing projection, or -a cross between -a stereoscope and the multiplex.
E. -PROCEDURES DESC3.IBsD BY SKIRIDOY
A, For Horizontal Control
-l. Ground triangulation
2. Ground traverse
3. Spatial triangulation, by
a. Universal Instruments
b. Stereometer
c. Multiplex
d. TsNI:IGAik (differentiated process)
e. Undistorted Model
f. Stereo-phototheodolite (Brief)
g. Mechanical triangulator (Brief)
4. Astronomical Stations
5. Radio Distance Methods.
Note: no mention is made here of our radial-line method, and apparently
it is not used. Geodeziya mentions also, and lays stress on, spatial tri-
angulation by measurement of coordinates on stereocomparator andnumerical
computation. This is probably the method of TsNIIGAMK. Except for Extension
Approved For Release 1999/09i02ETCIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 :RpP79-01083A000100040001-4
on Stereometer, all horizontal control methods used determine elevations
as well. Extension on Stereometer requires a line of levels down every
second side-overlap, but then it can determine elevations at the other
points.
B. For Vertical Control
1. Ground leveling
2. Barometric leveling
3. Straight-line method (most important)
4. Methods of Spatial Triangulation. See note above.
Note: Sometimes the straight line method is used to get the line of
levels required for Extension on Stereometer, and this then becomes a
complete bridging method. This is not accurate enough for large-scale
work.
0. For Auxiliary Control
1. Statoscope
2. Horizon Camera
3. Radio Altimeter
etc.
Note: Extent of use unknown.
D. Other methods -- for contouring, planimetry, orientation, etc., are
much like ours in the corresponding processes. There are some differences,
but-there is no point in going into this in detail. (See Geodeziya for
the Differentiated Processes Method.) Also, non-linear interpolation is
used when contours are interpolated from a few elevations.
L3
Approved For Release 1999/09/02 : DP79-01083A000100040001-4
Approved For Release 19 D2 : CIA-RDP79-01083A000100040001-4
P. GENERALS
The Method of Differentiated Processes uses a 600 C-factor. Whether
this is equiTalent to our type of C-factor or not is unknown, since the
accuracy of the contours (50% or 906) is not stated. The Photo Scales
used are:
Map Scale
Photo Scale
1:5,000
1:7,500 - 1:10,000
10,000
15,000 - 17,500
25,000
17,000 - 20,000
50,000
30,000 - 40,000
100,000
55,000 - 75,E
The straight line method is said to determine elevations to an accuracy
of H/700 to H/1100. This depends on measuring long lines to an accuracy
of .03 mm. to .05 mgt. I doubt that the instruments and materials described
can give that much accuracy.
Much mention is made of bridging, and it apparently is used extensive-
ly, but no experimental determinations of accuracy are given.
Accuracy of contouring by three methods was determined by a field teat
in 1945. Height error was related to average slope of terrain, with the
following results:
Let 4 be average terrain slope, m be standard error
in meters. Then:
Plane Table:
m - 0.8
15.0 tan a
Stereometer:
m = 1.3
4.4 tan a
Stereoplanigraph:
m = 0.8
1.2 tan a
44
Approved For Release 199-W91D2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : 1-DP79-01083A000100040001-4
Test was made on a 1:25,000 scale, 5 in. contour map.
The Russar-25 lens used in the ultra-wide angle multiplex is said to
have F = 20 mm., f/10, 122? field, resolution of 100 1/mm. at center,
60 1/mm. at edge. Every effort should be made to secure either an actual
lens or the drawings for the Russar-.22, 70 mm. lens and the Russar -25,
20 mm. lens..
45
Approved For Release 1999/09/O, IAA-RDP79-01083A000100040001-4
Approved For Release 199!RWR : CIA-RDP79-01083A000100040001-4
APPENDIX V '
ANALIS OF SOVIET PHOTOGRAMMETRIC INSTRL)MENTS
Based on Information Contained in a Book by
F. V. Drobyshev,
""Fotogrammetricheskiye Pribory i Instrumentovedeniye"
(Photogrammetric Apparatus and Instrumentation)
128k1A5a1
1. General Analysis
A review of Soviet photogrammetric instruments, based upon descrip-
tions in the textbook, '!Photogramraetric Instruments and Instrumentology*
by F. V. Drobyshev, Moscow, 1951, is generally disappointing from a tech-
nical point of 'riew. No new principles are disclosed, and the solutions
which have been selected for development into working models seem to be
more complicatedand less direct than those employed in western Europe and
the United States.
The mechanical and optical precision expected in Soviet instrument
design is apparently somewhat lower than in western Europe. The textbook
specifies that photographic image positions must be determined within
0,02 to 0.05 mm. Both Wild (Swiss) and Zeiss (German) require 0.01 mm
in their preciee instruments; Multiplex and Kelsh plotters used in the
United States give approximately 0.10 and 0,07 mm. respectively. Fur-
thermore, no mention is made of corrections for residual lens distortions;
25X1A5a1 western Europe manufacturers have given much thought to this problem, and
xhis analysis was originally prepared for a report entitled Soviet
Geodetic and Photogramnetric Instrumentation, CIA/RR-ER-5., 15 AAp -1951.
3TIMT. is repeate here for a convenience of readers to whom the
above report is not readily available.
46
Approved For Release 1999/09/02: CIA-RDP79-01083A000100040001-4
SECRET
Approved For Release 1999/09/02 sfRDP79-01083A000100040001-4
some method of correction is incorporated in each of the precise instru-
ments. Another indication of precision is found in the size of the mea-
suring marks. The Russian instrument designs call for marks of 0.05 to
0,,20 mm. in diameter; Wild specifies 0.04 mm (A5) and 0.02 mm. (A7 and
A8). One instance in which the Russians seem to have been ahead of de-
velopments in western Europe Is the use of luminous measuring marks of
various colors. These were not introduced by Zeiss until the Model C-7
Stereoplanigraph about 1948. They have still not been accepted by Wild
and Santoni (Italian). The Russians use them even in the simpler instruments.
The lower precision and the different emphasis in instrument design
may probably be attributed to the fact that the Soviet problem has been
largely the mapping of extensive areas at scales of 1:50,000 and. 1:100,000
and smaller. In western Europe the trend has been towards much larger
scale mapping with a consequent increase in precision requirements.
No credit is given anywhere in the book to any manufacturer of pho-
togrammetric.instruments outside of the U.S.S.R. A number of the instru-
ments described are exactly the same as Zeiss designs. It is most probable
that the descriptions are of the Zeiss instruments themselves. If not,
the instruments are direct copies of the Zeiss products.
This applies specifically to the following devices:
Small photo rectifier FTM
Large photo rectifier FTB
Horizontal stereocomparator
Radial triangulator
Stereoplanigraph C-4
None of the other western European designs has been adopted. Undoubtedly
the reason for this is that the Zeiss factory in Jena, Germany, is now
under Soviet control. The realization of space rays by mechanical rods,
47
Approved For Release 1999/09/02 : CUMDP79-01083AO00100040001-4
Approved For Release 199 2 : CIA-RDP79-01083A000100040001-4
successfully employed by Wild and Santoni, has not been used, although it
is mentioned brie:?ly as a possible solution.
A more detailed analysis of the different types of instruments follows.
II. _1nstrum1_nts
1. Photo Rectifiers
Five different instruments are described. The first two of these
are merely enlarging and reducing projectors having no provision for-tilt-
ing. Automatic focus is obtained in the first in-strument by means of a
rhombic inverter and in the -second by a cam and follower. Both are extreme-
ly large instruments with limited applications.
The third instrument is a true rectifier but has a very limited
range. The adjustment of focal distances is obtained by a spiral cam con-
trolled by a foot disc. This solution undoubtedly does the job, but takes
up a great deal of-room.
The fourth and fifth instruments described are direct copies of
the Zeiss small rectifier, SEG IT, and large rectifier, SEG I.
Nothing comparable to the Bausch and Lomb fully automatic rec-
tifier is describedL.
2. Stereo comparators
Two stereccomparators are described. The first is exactly the
same as the Pulfrich-Zeiss stereocomparator, no longer :uannufactured.
In the second instrument the plane of the photographs has been
tilted to make observations more convenient for a seated operator. Also
the least reading has been reduced from 0.02 to 0.01 mm.
Approved For Release 1 X02: CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : 4RDP79-01083A000100040001-4
Stereocomparators were originally developed for work with terrestrial
photographs. They have never been in favor in this country and have been
largely replaced by semi-automatic plotting instruments in Vest European
countries.
They are used primarily for obtaining coordinate measurements for
analytical methods, for analytical radial triangulation, and for non-
cartographic applications such as astronomical measurements.
At present, Nistri in Italy and the Cambridge Instrument Company in
England manufacture stereocomparators. The precision of measurement in
each of these instruments, 0.01 mm, is the same as that claimed for the
Russian device, although the system used for obtaining these values is
different in each of the three. In the Russian instrument full length
glass scales read by means of a mechanical micrometer are used. Nistri
uses a precise grid superimposed on the photograph in the plate holders;
coordinates in each grid square are read by means of a spiral micrometer.
The British instrument uses a calibrated grid imposed on the photographs
in the camera when the exposure is made; coordinates in each grid square
are read by a micrometer drum. This automatically cor=pensates for film
shrinkage, etc. Both the Italian and the British instruments have full
circle rotation of the plate holders making them adaptable for radial
triangulation. This feature is apparently not provided on the Russian
instrument.
It is stated that the collimation system of observation was applied
by Drobyshev seven years before it was adopted in foreign countries. This
is doubtful since Pulfrich used tha principle in the first Zeiss Stereo-
comparator built in 1901.
1.9
Approved For Release 1999/09/02 : CI DP79-01083A000100040001-4
Approved For Release 19990
02 : CIA-RDP79-01083A000100040001-4
Ster-eopantometer Drobyshev
This instrument performs the same function as the Abrams Contour
-Finder, the-Fairchild Stereocomparagraph, the Zeiss Stereotop, the Nistri
Stereographometer, or any combination of mirror stereoscope and parallax
bar. The Russian and Zeiss instruments move the photographs with respect
to-stationary measuring marks and stereoscope; the others move the-stereo-
scope and marks with respect to stationary photographs. So far as results
are concerned, one -system is as good as the other. The first results in
a more compact but more complicated instrument; the second is simpler and
cheaper to construct. The Russian device introduces a luminous floating
mark, probably a needless refinement in such an instrument. Nistri has
an attachment for producing an orthographic projection at constant scale.
The latest model of the Zeiss has -a device-for approximately correcting
-the effects of snail tilts.
4. Stereometer Type Instruments
The three stereometer instruments described illustrate the re-
sults of a purely nationalistic development. No comparable instruments
a3ze produced in this country or in western Europe. All three designs are
an extension of the principle of the stereocomparator based upon two
premises:
1. Both photographs will be maintained in the same plane
regardless of tilts which may have existed at the time
of exposure.
2. The line of sight of the optical viewing system will
be kept perpendicular to the plane of the photographs.
When tilted photographs are observed under these conditions, the observed
values of horizontal (x or height) parallaxes and vertical (y or orientation)
Approved For Release 199MRM : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 ~ 4 RDP79-01083A000100040001-4
parallaxes will be in error. In order to correct these observed values
the photographs must be moved in the x direction, in the y direction, and
about a vertical axis not necessarily coinciding with the camera axis.
These corrections are applied by ingenious mechanical devices which auto-
matically introduce the required motions. In order to keep these devices
simple, certain assumptions are made -- primarily that the tilts do not
exceed three degrees. The settings of the correction devices are functions
of the normal orientation elements.
A complete solution by this system would require three correction
devices on each photograph. Apparently such an instrument has not yet
been constructed. The three instruments described give only a partial
solution, and further operations with the photographs are required in
order to obtain complete map information from them. The instruments un-
doubtedly perform the function for which they were desighed, but in view
of the limited information obtained from them, they seem to offer no
advantages over the instruments in use in the west. They are not adapt-
able for extension of control by bridging methods, and based on the sim-
plicity principle, it is soubtful that a high degree of accuracy could
be attained.
a. Topographic Stereometer of Drobyshev
The end product of this instrument is elevations of ground
points and or contour lines drawn by hand directly upon unrectified pho-
tographs. In this respect it is roughly comparable to the Zeiss Stereotop
or the Brock Stereometer, except that rectified photographs are used in
the Brock instrument. All of these devices are subject to the limitation
that each contour line is at a different scale.
Approved For Release 1999/09/02 : ( RDP79-01083A000100040001-4
Approved For Release 199WQW92 : CIA-RDP79-01083A000100040001-4
b. Stereometer of Drobyshev
This instrument is a larger and more precise model of the
Topographic stereometer. It is adapted to use either glass platen illumi-
nated from below or paper platen illuminated from above. The y parallaxes
are measured and the least readings of the correction devices are smaller
so that more precise values of the orientation elements are obtained. Con-
-tour lines are drawn by a pencil attached directly to the instrument rather
than by hand.
c. Kern Stereometer (Skiridov)
This instrument is designed solely to determine the elements
of relative orientation of a stereopair by elimination of the y parallaxes
at five points. Elevations of contour lines are not determined.
d. Instruments of Direct Optical Intersection
Those are projection type instruments operating essentially
on the same principle as the Multiplex and Kelsh ;Potters. Again no new
principles are disclosed, but several interesting innovations are described.
It is stated that stereoscopic vision is obtained by the use of polarized
projectors and spectacles. Western experiments with this system have not
been successful.
5.
Projectors
a. Double Projector of TsNIIGAik (Drobyshev)
This instrument was apparently designed primarily for re-
search in various means of viewing the stereoscopic model. It offers a cor-
rect solution but hs.s drawback for practical application. Variation between
camera and projector focal lengths and plate sizes results in different
Approved For Release 1999 9/02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 i jqi ,RDP79-01083A000100040001-4
scales for horizontal and vertical measurements. Absolute orientation is
obtained by tilting the base table which could make drawing inconvenient.
The two projector instruments cannot be used for bridging.
b. Double Projector DPD-2 of Drobyshev
The instrument is designed for producing large scale topo-
graphic maps from contact diapositives. The projectors are disposed in
a horizontal position and the relative orientation is performed by intro-
ducing rotations to mirrors placed in front of the objectives. A similar
scheme was used in a German instrument designed by gasser in 1915. Abso-
lute orientation is still obtained by tilting the drawing table with re-
spect to the fixed projectors.
The model may be viewed either by the anaglyph principle or
by means of the blinking method. At present only Nistri applies the
blinking method in a production instrument. Experiments have been con-
ducted in this country but there is no unanimous preference for one method
over the other.
In end product the instrument is comparable to the Kelsh
plotter, but in construction and operation it is much more complicated.
6. Soviet Multiplex
In photogrammetric principles and operation this instrument is
exactly the same as the instruments produced by Bausch and Lomb, Williamson,
and others.
Optically, the ultra-wide angle coverage, 122?, and the use of
aspherical condenser lenses are of great interest. A distortion free
objective of this angular coverage and the specified resolving power of
53
Approved For Release 1999/09/02 :sqRDP79-01083A000100040001-4
Approved For Release 199999902 : CIA-RDP79-01083A000100040001-4
60 lines per millimeter sloes not exist in western instruments. To be
significant this lens must be complemented with a camera objective of
equal angular coverage. It has not been found practical to mass produce
aspherical lenses of high quality.
7.. Instruments of Optical-Mechanical Intersection
There are many different solutions possible in this category.
Those selected by the Soviets for development into working models are the
same as those used in Western instruments. In pursuit of the national
trend towards keeping the photographs co-planar, a system is described in
which existing photo tilts are introduced by providing an adjustable joint
in the space rods. A similar system was described in Swiss patents 251686
and 262481 by H. 'Wild in 1948, but no instrument using this scheme has
been built.
a, Stereo Universal of Skiridov
This instrument, like the Kern Stereometer of the same designer,
serves solely to determine the elements of relative orientation of two
photographs by elimination of y parallax in five points. in this instru-
ment the photographs are tilted, while in the stereometer instrument the
parallaxes are corrected with the photographs maintained co-planar. The
use made of this :limited information is not described. It is probable
that it is used for the settings of other instruments in which the photo-
graphs are actualLy plotted. Another possible use is to aid in determ-
ining the positions of nadir points and isocenters, after which the pho-
tographs may be used inradial triangulation with results equal to those
obtained in space triangulation.
51.
Approved For Release 1999112 : CIA-RDP79-01083AO00100040001-4
Approved For Release 1999/09/02 :DP79-01083A000100040001-4
b. Radial Triangulator
The illustration, diagram of optical system, and description
are precisely applicable to the Zeiss radial triangulator.
c. Stereoplanigraph S P 3 and C-4 (Konshin)
The C-4 seems to be a description of the Zeiss Stereoplani-
graph C-4, while the S P B Is the Soviet copy of the same instrument.
The photograph of the instrument, the diagrams of the mechanical and op-
tical systems show only minor discrepancies from the Zeiss. However, the
Soviet model does employ luminous measuring marks, not incorporated by
Zeiss until the model C7.
d. Stereoscopic Universal Instrument RP-6 (Konshin)
This instrument is an attempt to reduce the complexity of
the stereoplanigraph to speed up map compilation. It is not a universal
instrument at the term is understood here, since it is adaptable for use
only with near vertical aerial photographs.
In principle the instrument is a combination of a K E K Plotter
and a vertical sketchmaster. The photographs are oriented in space by
means of angular and directional motions. The elements of relative ori-
entation obtained in the Stereo Universal and Kern Stereometer of Skiridov
are probably used for this purpose. The mirror stereoscope allows view-
ing of the entire model at once. Half silvered mirrors give the impres-
sion of the model projected upon the drawing table. Planimetry and contours
are drawn by hand on this projected model. The instrument could not be
expected to give a high order of accuracy.
Approved For Release 1999/09/02 sTRDP79-01083A000100040001-4
Approved For Release 199WOW02 : CIA-RDP79-01083A000100040001-4
IN 15 cq1 5aj1 AND USA
Despite the criticism given to the C-factor, it is a useful measure
of comparative mapping process efficiency provided, in making comparisons,
that
(a) equivalent measures of contour accuracy are used,
(b) the vertical accuracy requirement is controlling, and
(c) physical conditions of use (referring to terrain and
atmosphere) are similar.
As a practical measure, if these conditions are observed, the C-factor
shows approximately how much map is obtained for a given effort, including
not only the physical differences between mapping processes but also the
effects of varying and not measurable factors such as personnel efficiency,
haze, materials control, etc. As a technical measure, applied to some
theoretically perfect or standard conditions, the C-factor measures the
potential vertical -information content of the photograph when the informa-
tion-is to be usedby the process under consideration. In addition, when
applied as a techn9'.cal measure the C-factor is strictly equivalent to
other incomplete -technical measures such as precision of parallax measure-
ment, contouring "apread", etc. Superior -overall measures can be obtained
only by more inten.bive process analysis than has yetbeen reduced to practice.
We will use the C-factor as a basis of comparison of Soviet and
American mapping processes, on the basis of the limited information available.
56
Approved For Release I 999 X 02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 : OI DP79-01083A000100040001-4
.Considering the conditions for comparability stated above in reverse
order,
(c) we can assume that the conditions of use are similar, since
the figures to be compared refer to average practice over
the years and over the wholes of the two countries, and
there are no outstanding physiographic dissimilarities be-
tween them,
(b) we will compare for large scale maps, where the vertical
accuracy is controlling,
(a) we must examine the Soviet map specifications relating to
contours and attempt to determine what an equivalent con-
tour interval would be according to American specifications.
Skiridov states (p. 300) that for terrain slopes of less than 2?, the
maximum error shall be 1/3 contour interval; for terrain slopes of 2 0 to
60 the maximum error shall be 2/3 contour interval; for 1:25,000 . maps.
The question is, what does 'maxim= error' mean? This is usually, in
America, taken to mean the 09!00 error -- not more than $ of the points
are to exceed the limit.
The American standard for contour accuracy is that NO of the points
are to be within 1/2 contour interval. The ratio of '95?' error to '90%'
error is 1.2. Applying the factors to land of less than 2? slope (which
we will use for our comparison), for the same accuracy the contour inter-
val on the Soviet standard is ~/
1.2 x 1/2 f 1i 3 1.8
ties the comparable American standard contour interval.
Skiridov states (p. 294) that the maximum theoretical C-factor (U.S.S.R.)
is 600 for large scale maps and 1:25,000 maps in flat areas, and that ae-
tual C-factors are lower. Analysis of the photo scales used (p. 296 and
later) and focal lengths as given-later in the same chapter indicates
57
Approved For Release 1999/09/02 :SdtDP79-01083A000100040001-4
Approved For Release 199?R 2 : CIA-RDP79-01083A000100040001-4
actual C-factors of 300 to 400 (U.S.S.R.), in flat country at 1:10,000 and
1:25,000 maps. Converting these to American C-factors by use of the factor
1.8, we have for the U.S.S.R.:
Theoretical maximum C-factor: 1080
practical average C-factor: 540 to 720
These refer to mapping by stem?planigraph and by the Method of Differen-
tiated Processes.
Comparable methods in America have the following C-factors
Stereoplanigraph Multiplex Ke30h
and Brock
Theoretical Maxim= 2000 1204 1404
Average practice 1250 - 1504 600 - 800 1000
'Thus, it appears that the average Soviet methods, as used., are about con-
parable to lltt.].tip'lex work, and considerably less accurate than the best
American methods. It is believed that the chief cause of the lower ac-
curacy In Russian work, if it exist-s, lies in the photographic materials,
since
(a) the permissible error in contouring is only 1.3 times the
precisioa.of primary parallax measurement on the photograph,
according to Skiridov, and
(b) toleranciss on distortion of photographic materials, as
stated im `Geodeziya' , is considerably-greater than American
toleranciss.
It is probable thi-t the planimetric accuracy is comparable to American
practice.
58
Approved For Release 199%#W : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 BDP79-01083A000100040001-4
to STR&IGRT LIII M MD 07 S3 SCTIJSB
She following translation has bees selected for inclusion in this
technical paper for three reasons; rarely, because (1) the method described
is one of the sore original Soviet methods and (2) the method is an illus-
tration of the relationship of Soviet photograametric methods to the spe-
cific and peculiar needs of that nation gad (3) the method sight be of
some interest to those individuals interested in photogrammetric reconm is-
sance in this country.
the original paper ray be found in the reference book, de s ,
Tom IX, Chapter 4, p255,a1214 dated 1949. The translation was made by
25X1 A5a Nr. B. Kri janovskiy
The "straight line method of Romanovskiy, U.S.S.R., is based on the
theorem that for any straight line lying in the object space photographed,
there corresponds a straight line on the photograph. And inversely, that
for a straight line on the photograph there can correspond any line on the
land, provided a plane can contain this line and the perspective center.
In this case, such a line would not be imaged as a straight line on the
adjacent photograph, except when all its points lie in the same basal
plane, or when it is actually a straight line.
Let us suppose, that three points have been selected on the left
photograph, lying on one straight line: a1, b1 and cl; to these points
correspond three points on the land: A, B and C (Pig. 157).
59
Approved For Release 1999/09/02 : 1 'DP79-01083A000100040001-4
Approved For Release 1999/Q/ CIA-RDP79-01083A000100040001-4
Pie. 157
If all-three points A, B and C do not lie in the plane containing
both perspective centers, (do-not lie in the sane basal plans) or are not
located on the same spatial straight line, it is impossible to define a
plane through these three points and through the second (right) perspec-
-tine center of projection. Therefore images of these three points a2,
b2 and c2 would riot be located on one straight line.
When we draw a straight line through points a2 and b2, the deviation
of the point c2 from this line is because of the elevation of the point C
on the land above the line connecting A and B. To prove this we draw a
straight line through points A and B, and produce it until it intersects
at of with the perspective ray slci. As ground points A. B and C' are
6o
Approved For Release I 999 QR : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 ~,gj"DP79-01083A000100040001-4
located on one spatial straight line, their images a2, b2 and c,', on the
right photograph would also be located on one straight line.
The elevation of the point C' above point A can be easily determined
from the right triangle (Fig. 158), when the elevation of the point B
above the point A and the distances AB and AC' are ',known. Then
D
hC' _ A = hB _ AD2 ?
1
where Dl is the distance AB, and D2 is the distance AC'.
As the point C does not lie on the spatial straight line AB, the
point c2 does not coincide with the point c2. In other words the devia-
tion of the point c' from the straight line a2b2c2 occurs as a consequence
of the elevation of the point C above the point CO and is expressed by the
formula:
H
hC _C' bdpc
(12)
where Hct is the altitude of the air base above the point C', b is the
photo base above the point C on the scale of the survey, and Ape is the
difference in x parallax.
Fig. 158
Therefore the elevation of the point C above the point A is determined
by the formula:
61
Approved For Release 1999/09/02 : 4RDP79-01083A000100040001-4
Approved For Release 1999/Qq& : CIA-RDP79-01083A000100040001-4
D2 He ?pc
hC A = -A D1 + b
(13 )
Determination of the elevation of the point C by formula (13) is in-
convenient, as >everal values remain unknown (eg. D1, D2)-. Therefore this
formula is transformed into a different one namely:
AC = A + A + HdIHc (14)
G A B A Bfksin
1+(Q-1, -
A
In this fo:.mula AC is the elevation of the point C; AA is the eleva-
tion of the poihit A; HA is the elevation of the camera station above the
plane, containing A, HC and AC' - are the heights of the flight line above
the planes, containing points C and C' respectively; B is the air base -
T is. the angle between the straight line and the image of the air base,
Ap is the length of the perpendicular from the point c2 onto the straight
line a2b2c2.
Expression Q is determined by the ratio:
d
d (15)
1
where d2 is the distance a1c1 and d1 is the distance alb1.
Formula (1+) can be given in a more convenient form for calculation:
H jAp
A+b.sin +ap-4hg-t1hQ (16)
AC AA+
423 +1
where AhH and &d q are the corrective terms computed according to the fol-
lowing formulas:
62
Approved For Release 1999/ T CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 S 1RDP79-01083A000100040001-4
(2B-A H hC-A)hC1-C
A
2
h
ahq (+Q - 1 } $ ~
A
(17)
The method of successive approximations is used, computing first the
basic expression (14) or (16), and then the corrections and repeating when
there is significant relief.
Therefore the straight line method makes it possible, given the ele-
vations of two points on the left photograph, to calculate the elevation
of the third point, located on the straight line through the first two.
This solution can be extended through succeeding air-photographs and this
permits a significant reduction in the number of vertical control points.
When this method is used the procedure is as follows: Two adjacent
strips are taken, and carefully assembled. On one of the strips, in the
zone of the side overlap, a straight line is drawn between two points
whose elevations are known. The straight line between these points is
produced and points are marked in the area of the triple overlap, on this
straight line; these points must be marked and are then identified on the
contact print of the adjacent strip. It is necessary that these points
be imaged in the triple overlap of the adjacent strip; otherwise, other
points must be selected. The first air-photographs of both strips are
put in a stereocomparator or in a topographic stereoscope and are oriented
in such way that the straight line, passing through three points of one
phothgraph and two points of the other photograph, will coincide with the
Y axis of the apparatus.
63
Approved For Release 1999/09/02 : C OP79-01083A000100040001-4
Approved For Release I 999 *d' : CIA-RDP79-01083A000100040001-4
Successive :rusion of the floating marks of the stereophotogrammetrical
apparatus with identical points of both photographs enables the difference
of the paralla:c of the third point in respect to the first two, which
are fused with the Y axis of the instrument, to be measured. The first
point (Fig. 159) has a known elevation, and the elevation of the second
one is deduced according to the relief of the terrain.
Fig. 159
The ratio Q is computed from formula (15), from the mean measurements
of the distances dl and d2 from the first point to the second and third
points on both photographs. The values b-and H are determined from the
horizontal radial line plot by-the ratios:
H A =Ikfk
A .
.1l eh
b_blkfk
HA
(18)
(19)
where l is the distance between any two points of the same elevation, on
different sides of the principal point and approximately at equal distances
64
Approved For Release I 99 M2 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02~#TRDP79-01083A000100040001-4
from it; k is the denominator of the scale of the radial line plot; ~ l?
is the distance between the same points on the airphotograph (Fig. 160);
Fig. 160
Fig. 161
dhA is the elevation of the first point above the plane containing
the distance 11; b1 is the distance between the principal points of the
photographs on the radial line plot. Better results are obtained if bl
is measured between the nadir points, but to do this it is necessary first
to determine the elements of relative orientation. When the differences
of elevation of the points on the straight line is small, it is sufficient
to use the distance between the principal points.
The angle T (Fig. 161) from equation (16) can be measured with a
protractor.
It is difficult to get the photographs in such a way in the stereo-
scope that the first two points appear to be exactly along the Y axis of
the instrument; and this is not usually accomplished. Instead, after
making measurements of differences of x parallax between the second and
third points and the first one, a computation is made to bring the differ-
ence of parallax of the second point to zero using the expression
dp3 = blp3 - QPIp2 ,
(20)
Where Atp3 and 0'p 2 are the measured differences of parallax of the sec-
ond and third points, and ep3 is the value for point 3 which would have
65
Approved For Release 1999/09/02 :sQRDP79-01083A000100040001-4
Approved For Release 1999MSM : CIA-RDP79-01083A000100040001-4
been measured if the first two points had exactly coincided with the Y
axis of the instrument.
When the elevation of the third point is determined, the second air-
photographs of bcth strips are set up in the instrument in a similar way.
On these photographs two points are plotted, whose elevations are known,
and a fourth point, whose elevation is unknown. Similar measurements are
made as before and with the aid of formula (16) the elevation of the fourth
point is determined. Those operations are repeated forall other airpho-
tographs of both strips, including the last ones, on which the last point
of the straight line has a known elevation. The difference between the
known elevation of the last point and that determined by the straight
line method, mare:3 it possible to determine the error, which was made by
the somewhat arbitrary determination of the elevation of the second point.
A correction is then made to all the elevations determined of points lo-
cated along the straight line by the expression
bh L
bhi= LA i
r
n
(21)
where bhi - is they correction in elevation of a certain point i;
bhn - is the difference between the known and determined elevations
Li andLn
of the! last point on the straight line;
- are the distances from the initial point of the straight line
to the point i and to the last point.
Certainly such a method of adjustment is only approximate and does
not take in consideration systematic errors (eg. thoseproduced by the
distortion of the surveying lenses) in-the measurement of differences of
X parallaxes. Therefore, a diagram is usually constructed based on an
66
Approved For Release 192 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 82RDP79-01083A000100040001-4
analysis of the method when a sufficient number of known elevations are
used and the character of the accumulation of errors is determined. On
the strength of this diagram, corrections may be introduced in later
operations.
The straight line method is applied in regions with small difference
of elevations (on the average no more than 50 - 100 m. on a stereopair).
Vhen there are greater differences of elevation points are selected which
appear to lie approximately at the same elevation. For the scale of
1:100,000, distances between terminal points equal to 4 - 5 bases are
permitted. For the scale of 1:50,000, determination of elevations by the
straight line method is usually limited to one stereopair, and then not
in the side overlap. The accuracy of elevations by the straight line
method is determined by the formula:
4 3
where b is the error of one transmission and n is the number of transmis-
sions. The error of one transmission usually depends on the accuracy of
the determination of the difference in X parallax, and therefore:
b bp, (23)
where by at the present time can be accepted as equal to ? 0.05 mm.
The sine rulers of Drobyshev are often used for determining elevations
by the straight line method. These give a sufficiently accuracy result
and are very simple.
These rulers (Fig. 162) consist of two glass plates, each having an
opaque line down their centers. One plate is rectangular, and the other
has one edge cut at an angle of 60. To determine the difference of
67
Approved For Release 1999/09/02 : 6 RDP79-01083A000100040001-4
Approved For Release 1999, : CIA-RDP79-01083A000100040001-4
Fig. 162
parallax by the straight line method, the airphotographs are adjusted under
a stereoscope and the rectangular glass plate is located on one of the
photographs in such a way, that,all three points lie under the opaque line.
The line of the second plate is laid over the two initial points of the
second airphotograph. One sees therefore one spatial line, combined with
the two; initial points. As the third point does not coincide with this
line at this stage (does not lie on the same spatial straight line), it
is necessary to shift the sine ruler in a direction perpendicular to the
line. For this purpose a metallic scale with millimeter divisions is
layed along the sloped edge of the sine ruler. The scale reading on the
scale opposite the edge of sine ruler is recorded. Now the sine ruler is
moved along the metallic scale until the third point coincides with the
spatial line. The number of millimeters on the metallic scale, passed
during this shifting are counted. This number, divided by 10 and multi-
plied by sin 6? gives the value of the difference in X parallax.
In many cases these rulers can also be used for the measurement of
Approved For Release 1999/-G9 02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 :Sc4DP79-01083A000100040001-4
the difference of ,X parallax of points not located on one straight line.
For this purpose two sine rulers and two metallic scales are used.
In this procedure the airphotographs are oriented along the principal
point base line, and the lines of the sine rulers are fused with identical
images on the airphotographs, at positions perpendicular to the principal
point base line. The metallic scales are then placed along the sloped
lines of sine ruler. One of the rulers is monocularly shifted so as to
coincide with a second point. This shifting is measured on the metallic
scale. After this, the second sine ruler is stereoscopically fused with
the image of the second point on the second photograph.
The difference of both shiftings, divided by 10, and multiplied by
Sin 6? is equal to the difference of X. parallax between the first point
and the second one.
69
Approved For Release 1999/09/02 SARDP79-01083A000100040001-4
Approved For Release t "M9/02 : CIA-RDP79-01083A000100040001-4
BIBLIOGRAPHY
1. Veselovskiy, N. U.: Fotogrammetriya (Photogrammetry) Izd. geodezich-
eskoy i kartograficheskoy literatury, GU( pri S1 SSSR, Moskva, 1945.
432 pp.
2. Skiridov, A. S.: Stereofotogrammetriya (Stereo-Photogrammetry)
Geodezizdat. Moskva, 1951.
3. Kell', N. G.: Opredeleniyeelementov vneshnego orientirovaniya plano-
vogo -snimka (Determination of the Elements of Exterior Orientation of
a Vertical Survey), pp. 104-133.
Nauchno-issledovatel'skiy Institut Aerofotos"yemki. Sbornik Statey,
No. I, Aeros"yemka (Aerial Surveying), 1934; No. II, 1934; No. III,
1935.
4. Gapochko, G. F.: Aeros*yemka v melkikh masshtabakh (Aerial Surveys
-of Small Scale). Trudy Gosudarst_xennogo Nauchno-issledovatel'-skogo
Instituta Geodezii i kartografii, Vyp. IV, 1931,'PP. 65-72.
5. Drobyshev, F. V.: Stereometr drobysheva (The Drobyshev Stereometer)
Nauchno-Issledovatel)skiy Institut Aerofotos"yemki. Sbornik statey
III, Aeros"yemke (Aerial Surveying), 1935, ppp? 59-67.
6. Drobyshev, F. V.: Stereometr i yego razvitiye (The Stereometer and
Its Development). Geodezist, No..11, 1936, pp. 62-71. Geodezist,
No. 12, 1936, pp. 37-45.
7. Drobyshev, F. V.: Topograficheskiy stereometr (Topographic Stereometer).
Geodezist, No. 6, 1937, pp. 10-17.
8. Rusinov, M. M.: Novyy shirokougol'nyy ortoskopicheskiy-aerofotoob"yektiv
Liar-6. (New Wide-Angle Orthoscopic Aerc-Photo Objective, the Liar-6)
Nauchno-Issledovatel'skiy Institut Aerofotos"yemki, Sbornik statey III,
Aeros"yemka (Aerial Surveying), 1935, pp. 38-58.
9. Nastavleniye po topograficheskoy s"yemke v masshtabe 1:100,000, Chast' II,
Fotogrammetricheskiye raboty. (Instructions for Topographic Surveys
on a 1:100,000 Scale, Part I~aI~~ ~-P",hotogr-ammetric Operations) Second
Edition, 1950. 277 pat&X1 hoye Upravleniye Geodezii i Kartografii
pri Sovete Ministrov.
10. Konshin, M. D., Rusinov, M. M., Yutsevich, Yu. K. and Sokolova, N. A.:
Geodeziya (Geodesy), Tom IX, 1949, Glava (Chapter) 4. "Small Scale
Aerial Topography,' pp. 205-292.
Approved For Release ' 99Q9/02 : CIA-RDP79-01083A000100040001-4
Approved For Release 1999/09/02 :tDP79-01083A000100040001-4
11. Drobyshev, F. V.: Potogrammetricheskiye pribory i instrumentovedeniye
(Photogrammetric Apparatus and Instrumentation) Izd. good. i Kartogr.
literatury, 1951-
12. After, S. P.: Opyt Kameral'nogo deshifrirovaniya aerosnimkov pri
Kartografirovanii v masshtabe 1:100,000 (Experiment in Office Inter-
pretation of Aerial Surveys for 1:100,000 Scale Mapping) Sbornik
nauchno-tekhnicheskikht i proizvodstvennykh statey po geodezii, karto-
grafii to o rafii aeros" emke i ravimetrii V . XVIII, . 77-81,
1948. 25X1A5a1
13. Kell', N. G.: Primeneniye azimutal'nykh proyektsiy dlya resheniya i
issledovaniya fotogrammetricheskikh ssdaah (The k-aplication of Azimuthal
Projections for Solution and Investigation of Photogrammetric Problems)
Izvestiya Akademii Nauk SSSR, seriya geograficheskaya i geofizicheskaya,
Tom XII, No. 2, pp. 119-136, 1948.
14. Perkis, L. I.: Postroyeniye planovykh fotogrammetricheskikh setey
grafo-analiticheskim sposobom (Construction of Vertical Photogramme-
trical Networks by the Grapho-Analytical Method) Sbornik nauchno-
tekhnicheskikh i proizvod-stvennykh statey po geodezii, kartografii,
aeros"yemke i gravimetrii, Vyp. XXIV, pp. 27_35, 1949.
15. Shevchenko, F. P.: Sovmestnaya uvyazka planovykh fotogrammetricheskikh
setey na razrezhennom geodezicheskom obosnovanii (Simultaneous Closure
of Vertical Photo;reamnetric Networks on the Basis of Sparse Geodetic
Control) Sbornik nauchno-tekhnicheskikh i proizvodstvennykh statey
po geodezii, kartografii, aeros"yemke I gravimetrii, Vyp. XXVIII, 1948,
pp. 46-52.
16. Durnev, A. I.: Vidoizmenennyy sposob zasechek (Modified Method of
Fixes). Sbornik nauchno-teldinicheskikh i proizvodstvennykh statey
po geodizii, kartografii, topografii, aeros"yemke i gravimetrii, Vyp.
XIII, pp. 49-60, 1946.
17. Drobyshev, F. V.: Perspektivnaya stereometricheskaya zasechka (Per-
spective Stereometric Intersection) Sborn.ik nauchno-tekhnicheskikh
i proizvodstvennykh statey po geodezii, kartografii, topografii,
aeros"yemke i gravimetrii, Vyp. XIII, pp. 21-34; 1946.
18. Volosov, D. S.: Methdy rascheta slozhnykh fotograficheskikh sistem
(Methods of Calculation of Complex Photographic Systems). Gosudarst-
vennoye izdatel'stvo tekhn.iko-teoreticheskoy literatury, 1948, Moskva.
Approved For Release 1999/09/02 :sCMRDP79-01083AO00100040001-4
Approved For Release I 99 CIA-RDP79-01083A000100040001-4
- -RET
Approved For R P79-01083A000100040001-4