MULTI-SENSOR IMAGERY RESEARCH: (AROUND-THE-CLOCK AERIAL SURVEILLANCE OF TARGET ACTIVITY LEVELS IN A SUSPECTED MILITARY BUILD-UP)

• Document Type: CREST
• Collection: NGA Records (Formerly NIMA)
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78B04747A001000020033-0
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 9
• Document Creation Date: December 28, 2016
• Document Release Date: April 27, 2011
• Sequence Number: 33
• Publication Date: February 10, 1964
• Content Type: REPORT

Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
? 1 _Kt1 0
10 February 1964
MULTI-SENSOR IMAGERY RESEARCH: (Around-the-Clock Aerial Surveillance of
Target Activity Levels in a Suspected Military Build-up)
1. INTRODUCTION.
(PHASE I)
Research is proposed to determine the most effective techniques of usin
sequential day and night imagery (from a combination of c era and
scanner) to determine changes in target activity. More s ecifically, the,
proposed project would seek to simulate acquisition and a alysis of changing
activity levels of certain targets which are indicators o military "build-
up." It should be emphasized that this research is not directed toward imagery
acquisition methods, but to the problem of imagery analys s and exploitation
techniques. The acquisition phase of this project would provide imagery taken
under rigidly controlled conditions -- to be used in developing objective ex-
25X1 ploitation techniques.
0 sensors have been flown over numerous targets to determine sensor
effectiveness in general-purpose detection and identification. These studies
25X1 have shown that, scanners are capable of acquiring imagery which
has sufficient resolution 'too-identify most targets of military interest. None
of the projects, however, have demonstrated that can be
25X1 effectively related to photography obtained a fe ' n~-o e r ier ow r, it
now appears that this could be one of the more obvious us s of =scanners in
the near future: for correct analysis of daylight photo raphy plus
over the same targets would be an ideal method of con-
tinual, around-the-clock surveillance of an area of suspected military "build-
up." Yet, to date, there is little objective knowledge about the quantity or
reliability of data which can be interpreted from ^ imagery under these
circumstances. Furthermore there is little knowledge about the additive
intelligence value of the "history" of a suspect target over a period
? of a week or so. It is suspected that activity-level inte pretation can be
improved considerably by this additive informatio ,
2. CONCEPT.
imagery and complete ground truth of seven targets over a eriod of seven
days and nights,
activities, as listed below:
2.2.1. Data Acquisition: Collection of aerial photog aphy,
it
2.1. Purpose. There are two primary objectives of th proposed research:
2.1.1. To determine the significance and value of in augmenting
conventional aerial photography as they both might be used for determining the
location and rate of military activity over a period of time;
2.1.2. To devise guidelines that will assist interpreters in their use of
sequential photo for military "build-up" analyses.
2.2. Scope. This project will encompass several sepa ate but coherent
SECRET
GROUP I
E.cI.d.d I,,,,
d..ngr.d~nq
Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
 Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
? ~~\..RC 1 ~ ill
2.2.2. Imagery Interpretation: Detailed interpretation of targets' activity
on aerial photography alone and then on photography and imagery in the
sequence flown, but without ground truth.
2.2.3. Data Analysis: Comparison of interpreted data with ground truth data
to determine the significance and additive value of infrared for target activity-
level analysis.
2.2.4. Methods Development: Derivation and presentation of guidelines to
assist interpreters of sequential imagery in their analysis of target activity
levels and military build-up.
There is strong opinion that the tasks required for the data acquisition
phase of this project should be performed separately and b a group other than
the one performing the subsequent interpretation, analysis,, and methods de-
velopment tasks. This will not only assure complete objectivity of conclusions
but will more nearly simulate actual operational conditions. Therefore, the
imagery interpretation, data analysis, and methods development phases will be
performed in-house and/or by a separate contractor. Only those requirements
relating to the data acquisition phase are listed below.
3. SPECIFICATIONS.
3.1. Data Acquisition Requirements.
3.1.1. Aircraft. The contractor shall provide an aircraft and crew
capable of fulfilling the project requirements as stated below.
3.1.2. Aerial Photography. Complete vertical photographic coverage of
each target area shall be obtained between 0900 hours and,1500 hours (local
time) each day for seven consecutive days.
The camera used must be a high quality reconnaissance or cartographic type
capable of providing large scale (1:5000 or larger) high quality photography.
Since stereo is essential, 60% forward overlap is a rigid requirement. If
more than one flight-line is necessary to cover a target, !side-lap coverage
should be at least 10% of the format width.
A ground resolution target shall. be photographed each day at some point of
that day's flight.
The film shall be a high-quality panchromatic emulsion. A gray-level step
wedge shall be attached to the end of each roll of unexposled film as an aid
for calibrating subsequent processing. Film shall be processed according to
? high-quality laboratory procedures. One complete set of negatives, duplicate
25X1 positive transparencies and paper prints shall be delivereld. All photography
shall be adequately identified and titled with target, date, time and altitude
of acquisition. Accurate flight-line plots shall accompany all photography.
SECRET
? Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
 Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
SECRET p
3.1.4. Ground Truth Data. In order to fulfill the research objectives, it
is extremely important that complete, accurate, and fully documented data on the
actual target conditions be gathered at the time of imagery acquisition. It is
realized that collection of the necessary ground truth isj a difficult task and
requires considerable personnel commitment; however, this data is probably the
most essential information for objective analysis of the total problem. Listed
below are the required essentials of ground truth:
3.1.4.1. Complete U. S. Weather Bureau hourly reports from the weather
station(s) nearest the area of targets. These reports s all cover a period
10from 24 hours prior to the first infrared mission throug , and including, the
last infrared mission. They should include at least: ai temperature, preci-
pitation, wind, cloud cover present, cloud levels and dev point. Also to be in-
cluded are the times of official sunset and sunrise for each day of operation.
3.1.4.2. Target temperatures with radiometers. Thig would include the
surface temperatures of selected target components, background surface tempera-
tures adjacent to measured target components, and ambien air temperatures.
These temperatures should be measured for the seven nights at approximately the
same time as the overflight in each target areal(plus or minus one hour).
25X1 3.1.4.3. Flas photographs and description of target elements which have
25X1 changed since the last mission.
3.1.4.4. Port Authors y togs, airport logs, and oth r similar target data
which is normally maintained on the status of the various target areas.
3.1.4.5. A full analysis of all exposed materials in each target area.
This should be accomplished before imagery acquisition since this data would be
quite useful in planning the location of subsequent temperature measurement.
? These analyses should result in seven target maps (color-coded or zip-coded) to
indicate the different surface materials in each target rea. ("Materials" are
defined as: concrete roof, aluminum roof, sod, grass, so 1, blacktop road, water,
etc.)
3.1.4.6. Complete documentation of all ground truth data which has been
collected shall be presented in the final technical report along with a descrip-
tion of the acquisition of all photography and imagery. All parameters
of each flight of photography and imagery -- e.g., time Over each target, alti-
tude, etc. -- should be identified. A complete description of the technical
characteristics of the camera and the scanner, as well as a description
of how each roll of film was developed and reproduced, siiall be included. Three
copies of the final Technical Report are required,
25X1 3.1.4.7. All photography, 0 imagery, and ground truth data shall
become the property of the United States Government.
3.1.5. Targets. There are seven target types which, are considered as
representative of military "build-up" indicators. The t~rget types are:
3.1.5.1. a port and its associated facilities
3.1.5.2. a large railroad yard
SECRET
? Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
 Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
? 3tLRt1 46
3.1,5.3, a civilian airport
3,1.5,4. a storage depot
3.1.5.5, a trucking terminal
3.1.5.6. a military motor pool
3.1.5,7. an unspecified facility or installation under rapid construction
Proposals for the data acquisition should include recommendations of specific,
existing installations of the above types. Final selecti n and delineation of
installations to be used in this project shall be approve by the Government
Project Monitor. For maximum efficiency of aircraft time ',and ground truth col-
lection, the selected installations should each be characteristic of their type
yet compact in area. And, it would be advantageous if they should all be in the
same approximate area (such as in one metropolitan area).' They should also be in-
stallations with some degree of activity around-the-clock;-- preferably with
changing rates of activity.
?
SECRET
? Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
 Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
JtLNt 1
10 January 1963
MULTI-SENSOR IMAGERY RESEARCH: (Around-the-Clock Aerfial Surveillance of
Target Activity Levels in a Suspected Military Build-up)
1. INTRODUCTION.
Research is proposed to determine the most effective techniques of
using sequential day and night imagery (from a, combination of camera and
25X1 to determine changes in target activity. More specifi-
cally, the proposed project would seek to simulate acquisition and analysis
of changing activity levels of certain targets which are indicators of mili-
tary "build-up." It should be emphasized that this research is not directed
toward imagery acquisition methods, but to the problem of imagery analyses
and exploitation techniques. The acquisition phase of this project would
? provide imagery taken under rigidly controlled conditions -- to be used in
develo bjective exploitation techniques.
25X1 sensors have been flown over numerous targets to determine
sensor effectiveness in general-purpose detection and identification. These
25X1 studies have shown that, at night, 0 scanners are capable of acquiring
imagery which has sufficient resolution to identify most targets of military
interest. None of the projects, however, have demonstrated that night F_
uses of ^ scanners in the near future: for, correct analysis of daylight
25X1 photography plus night time imagery over the same targets would be
an ideal method of continual, around-the-clock surveillance of an area of
suspected military "build-up." Yet, to date, there is little objective
25X1 knowledge about the quantity or reliability of data which can be interpreted
from [::]imagery under these circumstances. Furthermore, there is little
knowledge about the additive intelligence value of the "history" of 25X1
a suspect target over a period of a week or so. It is suspected that
? activit -level interpretation can be improved considerably by this additive
information.
earlier. However, it now appears that this could be one of the more obvious
research:
2.1. Purpose. There are two primary objectivesi of the proposed
2.1.1. To determine the significance and value of in aug-
menting conventional aerial photography as they both might be used for
determining the location and rate of military activity over a period of
time;
2.1.2. To devige guidelines that will assist interpreters in their
use of sequential photo/0 for military "build-up" analyses.
2.2. Scope. This project would encompass several separate but co-
herent activities,. as listed below:
SECRET
GROUP I
E.cl.d.d from_omo,<
do.ngra - and d e clo.i~lico~~o~
? Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
 Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
0 JCI.KC 1 0
2.2.1. Data Acquisition: Collection of aerial photography, I
imagery and complete ground truth of seven targets over a. period of seven
days and nights.
2.2.2. Imagery Interpretation: Detailed interpretation of targets'
activity on aerial photography alone and then on photography and
imagery in the sequence- flown, but without ground truth.
2.2.3. Data Analysis: Comparison of interpreted data. with ground
truth data to determine the significance and additive value of
target activity-level analysis.
2.2.4. Methods Development: Derivation and presentation of guidelines
to assist interpreters of sequential imagery in their analysis of target
activity levels and military build-up.
There is strong opinion that the tasks required, for the data acquisi-
tion phase of this project should be performed separo.tely and by a. group
other than the one performing the subsequent interpretation, analysis, and
methods development tasks. This will not only assure complete objectivity
of conclusions but will more nearly simulate actual operational conditions.
It could be accomplished by two individual contractors or a contractor-sub-
contractor arrangement -- implying that parties interested in participation
? should select either the acquisition phase or the three subsequent tasks
and submit proposals accordingly.
25X1 A good scanner is not available to thi$ sponsor for issue as
GFE for the research. It is therefore assj.med that potential bidders for
25X1 the acquisition of airborne and ground data will hav available a state-of-
the-art F__-Iscanner for use on this project.
3. REQUIREMENTS
3.1. Data Acquisition Requirements.
3.1.1. Aircraft. The contractor would providelam_ aircraft and crew
capable of fulfilling the project requirements as stated below.
3.1.2. Aerial Photography. Complete vertical photographic coverage
of each target area shall be obtained between 0900 hours and 1500 hours
(local time) each day for seven consecutive days.
? The camera, used must provide photography with b?th large scale (1:5000
or larger) and wide lateral coverage (800 or more). I70mm format is prefer-
able but consideration of scale, altitude, focal length and field angle of
the lens would make larger formats acceptable.
Since stereo is essential, 60% forward overlap is a rigid requirement.
If more than one flight-line is necessary to cover ai,target, side-lap
coverage should be at least 10% of the format width.''
A ground resolution target shall be photographed each day at some point
of that day's flight.
The film shall be a high-quality panchromatic emulsion. A gray-level
step wedge shall be attached to the end of each roll', of unexposed film as
an aid for calibrating subsequent processing. Film hall be processed
according to high-quality laboratory procedures. 0n complete set of nega-
tives, duplicate positive transparencies and paper prints shall be delivered.
All photography shall be adequately identified and titled with target, date,
time and altitude of acquisition. Accurate flight-lne plots shall accom-
pany all photography.
SECRET
Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
 Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
9
3.1.4. Ground Truth Data. In order to fulfill the research objectives,
it is extremely important that complete, accurate, and fully documented data
on the actual target conditions be gathered at the time of imagery acquisi-
tion. It is realized that although collection of the necessary ground truth
is a difficult task, this data is probably the most e~sential information
for objective analysis of the total problem. Listed below are the required
essentials of ground truth:
3.1.4.1. Complete U. S. Weather Bureau hourly reports from the weather
station(s) nearest the area of targets. These reports shall cover a period
from 24 hours prior to the first mission through, and including,
the last mission. They should include at least: temperature,
3.1.4.3. Flash photographs and description of target elements which
have changed since the last mission.
3.1.4.4. Port Authority logs, airport logs, and other similar target
data which is normally maintained on the status of th various target areas.
3.1.4.5. A full analysis of all exposed materials in each target area.
This should be accomplished before imagery acquisition since this data would
be quite useful in planning the location of subsequent temperature measure-
ment. These analyses should result in seven target maps (color-coded or
zip-coded) to indicate the different surface material in each target area.
("Materials" are defined as: concrete roof, aluminum roof, sod, grass, soil,
blacktop road, water, etc.).
Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
 Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
3.1.4.6. Complete documentation of all ground truth data which has
been collected shall be presented in the techn' a retort alon with a des-
cription of the acquisition of all photography All
parameters of each flight of photography and imagery -I,- e.g., time over each
target, altitude, etc. -- should be identified. A complete description of
the technical characteristics of the camera and the scanner, as
well as a description of how each roll of film was developed and reproduced,
shall be included.
3.1.4.7. All photography, and ground truth data
shall become the property of the United a es overrmjent.
3.1.5. Targets. Specific targets for this research will not be selec-
ted until initiation of the project. There are seven 'target types, however,
which are considered as representative of military "bi4ild-up" indicators.
Since these types are rather common in many areas of the United States,
selection of specific installations can be made after the contract is awarded.
Maximum efficiency in use of aircraft and personnel would be attained if the
specific targets were near the contractor's base of operations. The target
types are:
3.1.5.1. a port and its associated facilities
3.1.5.2. a large railroad yard
3.1.5.3. a civilian airport
3.1.5.4. a storage depot
3.1.5.5. a trucking terminal
3.1.5.6. a military motor pool
3.1.5.7. an unspecified facility or installation} under rapid con-
struction.
Proposals for the data acquisition phase should include recommendations
of specific, existing installations of the above types. For maximum effi-
ciency of aircraft time and ground truth collection, the selected instal-
lations should each be characteristic of their type y4t compact in area.
And, it would be advantageous if they should all be in the same approximate
area (such as in one metropolitan area). They should 'also be installations
with some degree of activity around-the-clock -- preferably with changing
rates of activity.
3.2. Imagery Interpretation Requirements.
. 3.2.1. Photographic Interpretation. A detailed!interpretation of all
daylight aerial photography of all targets shall be ccynducted without refer-
25X1 ence to ground truth or imagery. This task shall be performed in.
the same sequence as the photography was acquired and will consist of a com-
plete inventory of all elements, within the defined target areas, which are
indicative of activity levels of the installations. For example, the inter-
pretations of a port should include the number and types of ships in port --
their location and types of cargo on board, if evident; number and types of
off-loading equipment; amount and type of material inloff-loading areas; etc.
Thus, the interpretation of this port on seven days o photo coverage should
yield a history of daylight activity for that week. All interpreted data
should be tabulated so that comparisons can easily be,made from day to day.
In addition ground resolution of all photo coverages 4hould be measured from
the imaged resolution target.
3.2.2. Sequential Interpretation. This task is,essentially the same
as the previous photo interpretation except that all overage (photo and
will be interpreted in sequence of time acquired. (This is also to be
accomplished without ground truth data.) In order to~be as objective as
possible during this phase, the photography should be~re-interpreted in
SECRET
Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
 Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0
proper sequence- with the ~ without benefit of the results of previous
photo interpretation: the interpreted data may possibly influence
succeeding photo interpretation of the target. All interpreted inventory
data shall be tabulated so that activity is easily comparable on a day-to-
day basis. Ground resolution of the imager. should be closely
zaAI estimated on all ocoverages.
3.3. Data Analysis Requirements. Analysis of accumulated data at this
stage of the project should be designed to satisfy tYree objectives:
3.3.1. to determine the relative intelligence Outputs of "day only"
interpretation vs day and night interpretation. This can be achieved by
comparing the initial photo interpretation data, agairst the sequential
25X1 photo data;
3.3.2. to determine the total effectiveness of night for
inventory of activity level indicators. This can be'accomplished by a com-
25X1 pa.rison of interpreted data, with ground truth;
3.3.3. to determine the intelligence significance of the thermal
"histories" of certain target elements.
It is hoped that this data analysis phase can be so designed that the
S results can be compared as objectively as possible. However, since the seven
targets are dynamic in nature and will undergo many changes over a period of
a week, in many cases strict comparison of sequential imagery of the same
objects will not be possible. In these cases, it is ''probably valid to sub-
jectively compa.re imagery of similar objects on a "best judgment" basis.
All data and the results of data analysis should be incorporated into an
interim technical report.
3.4. Methods Development Requirements. As a, result of the knowledge
gained during the interpretation and analysis phases,' an interpretation
manual shall be produced. This manual should be suitable as a training and
reference source for qualified photographic interpreters who have little
knowledge or experience with night imagery. Accordingly, it should
include an introductory section on simplified physics,
equipment and on the characteristics of
imagery in general. The
main section of the manual shall consist of guidelines for interpreting
25X1 activity levels of build-up indicator targets on sequential day and night
photography and imagery. The final design, ,content, and format of
the manual shall be subject to approval of the Government contract monitor.
Fifty (50) copies of this manual are required. All continuous-tone
illustrations shall be photographically reproduced soy as to retain maximum
resolution and clarity.
SECRET
Sanitized Copy Approved for Release 2011/04/27: CIA-RDP78BO4747AO01000020033-0