RESEARCH AND DEVELOPMENT PROJECT APPROVAL REQUEST

• Document Type: CREST
• Collection: NGA Records (Formerly NIMA)
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78B04747A002600020013-8
• Release Decision: RIPPUB
• Original Classification: C
• Document Page Count: 2
• Document Creation Date: December 28, 2016
• Document Release Date: May 25, 2011
• Sequence Number: 13
• Content Type: REQ

Sanitized Copy Approved for Release 2011/05/25: CIA-RDP78BO4747AO02600020013-8
VMW
RESEARCH AND DEVELOPMENT
PROJECT APPROVAL REQUEST
I. Identification
The Development Branch of the Plans and Development Staff proposes
a study to determine the feasibility of developing an "Automatic Focusing
System". The project will be conducted
level and is programmed for Fiscal Year 1965 under Category I, Special
Techniques and Development Studies". Within this category, it is pro-
grammed under "Semi-automatic Photo Interpretation Techniques".
II. Objectives
This project would prove feasibility of an automatic focusing system
for potential development. Essentially, the program would establish the
feasibility of: (1) detecting image sharpness by using the autocorrelation
of the image as an indicator of sharpness; (2) using a solid-state correla-
tor as the image sharpness sensor; and, (3) developing a signal for driving
the projection lens to the position of best focus. The study is intended
to show that an automatic focusing system, which can be linked to present
projectors by the appropriate servo loops, can be produced. The automatic
focusing system will consist of the solid state correlator for obtaining
the autocorrelation function of the image; an optical system for projecting
two identical images on the sensor; and the required electronic circuits
for converting the autocorrelation signals into a form for controlling
a servo motor to, in turn, drive the projection lens to the position of
best focus. Moreover, the study would determine sensitivity of the auto-
correlation function to focus and, in particular, the sensitivity of the
developed correlation sensor. ffort should yield designs
for a compact, economical system for achieving and maintaining focus on
various projection and direct-viewing devices?
III. Background
Correlation systems now in use generally require a scanning system,
an electronic multiplier, and an electronic integrator -- together with
the associated power supplies, deflection circuits, focusing circuity,
amplifiers, etc.
Such equipment is too costly, and complex for automatic focusing
with typical reconnaissance viewers. Other systems in use, such as
I
25X1
CONFIDENTIAL Idow "I'll
Sanitized Copy Approved for Release 2011/05/25: CIA-RDP78BO4747AO02600020013-8
 Sanitized Copy Approved for Release 2011/05/25: CIA-RDP78B04747A002600020013-8
mechanical linkage devices, require periodic calibration, and these
systems are designed to be used with specific lens systems. It would
be extremely beneficial to develop a system that would not require
periodic calibration and could be applied to a variety of lens systems.
has submitted an unsolicited proposal to develop
25X1
such an automatic focusing system. Because the corporation has applied
for patents for their system, they must be considered a sole source
supplier. , is considered to have the technical capability
25X1
to develop the system outlined in the technical specifications.
Technical Specifications
Basically, a correlator can be used to sense image sharpness
because the cross-correlation of two images, both in sharp focus,
is substantially higher than the cross correlation of two defocused
images. Plotting light intensity vs position, it can be shown that
a focused image will have a higher value on the intensity scale at a
given point than the defocused image. This is because the total
energy under the curve must be conserved. In the defocused case,
the light is spread over a wider position so the intensity must be
lower. The correlation function is obtained by multiplying the
intensities of the image, point by point and adding the contributions
from all the image points. It is clear that the peak of the corre-
lation curve is lower for the defocused image.
The use of the correlation technique is not new but its appli-
cation to automatic focusing has been limited because of the cost, size,
weight and power required by the implementation. With the use of the
solid state correlator developed by it should be possible 25X1
to implement an automatic focusing system that would overcome the
problems cited above.
V. Contract and Financial Arrangements
cost-type contract
for the demonstration of
feasibility, design and construction of a breadboard solid-state
automatic focusing system. Delivery of the final report is scheduled
for three months after the date of award. of contract. (This contract
is recommended on a sole source basis because of the proprietary sensor
proposed for use in the development of the system.)
Security
25X1
25X1
CONFIDENTIAL
Sanitized Copy Approved for Release 2011/05/25: CIA-RDP78B04747A002600020013-8