ROTATING PRISM SCANNER

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP79B00314A000500040003-0
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 6
• Document Creation Date: December 19, 2016
• Document Release Date: December 13, 2005
• Sequence Number: 3
• Publication Date: January 20, 1964
• Content Type: REPORT

Approved F1elease 2005/12/23: CIA-RDP79B003 000500040003-0
20 January
Page 1
ROTATING PRISM SCANNER
The rotating prism scanner is a device employing photoelectric setting on
an image. Our attention was directed toward such a device as a potential aid
in eliminating some of the subjective variation involved in RES measurements.
It is claimed that the device substantially enhances the reproducibility in
measurement of a given image by a given operator over "eyeball" techniques,
and further, that the variation from operator to operator is substantially
reduced.
Basically, the device operates as follows. Light passing through the film
or plate containing the image of interest impinges on a beam splitter. Part of
the light passes to a viewing screen, where the magnified image is displayed.
The other portion of the light passes through a rotating prism to a slit, be-
hind which a photomultiplier is located. The rotating prism causes the image
of interest to be swept across the slit. The output of the photomultiplier is
applied to the Y axis of an oscilloscope. A rotating sector or chopper is
physically attached to the same shaft as the prism, so that perfect synchroni-
zation is achieved. The output of a photocell or photomultiplier, illuminated
through the sector or chopper, is applied to the X axis of the oscilloscope.
The configuration of the sector or chopper is such that the following sweep
sequence results. As the scope is sweeping from left to right the transmittance
of the line or edge under study is displayed from left to right. After a hold
period the scope sweeps from right to left, displaying the same curve, but now
in an opposite direction. The result is a display of the curve and its mirror
image. For example, the display of a symmetrical image such as a spectral line,
would appear as in Figure 1.
XEROi
neu,
XEF
C OF
Approved For Release 2005/12/23 : CIA-RDP79B00314A000500040003-0
 Approved F
Figure 1.
An unsymmetrical image, such as an edge, might appear as in Figure 2.
Figure 2.
Approved For Release 2005/12/23 : CIA-RDP79B00314A000500040003-0
 Approved Foolease 2005/12/23 : CIA-RDP79B003W000500040003-0
I I STAT
anuary
20
Page 3
The two traces can be moved relative to each other by movement of the
carriage supporting the plate or film. The precision of the device depends
upon the sensitivity of the eye and brain to a symmetrical situation. For
the spectral line in Figure 1, the position of the center of the line is read
when the two curves coincide.
The measurement of the width of an edge is more complex. However, this
. en?t /flJ.s7_
width can be defined, and reproducible measurements obtained. For example,
the height of the traces in Figure 2 can be brought to a standard value between
two reference points by amplifier adjustment. Then the first reading might be
taken when the two traces intersect at, say, 85% of maximum amplitude, as shown
in Figure 3A. The final reading might be taken when the 85% amplitude point
has been moved to the same abscissa as the previous 15% amplitude point, as in
Figure 3B.
Figure 3A
XE'
cc
Approved For Release 2005/12/23 : CIA-RDP79B00314A000500040003-0
 Approved Fc~lease 2005/12/23 : CIA-RDP79BO03W000500040003-0
Page 4.
The edge width, as defined, is then the difference between the two
readings.
It is apparent that such a technique allows reasonable reproducibility
with a given edge for a given operator and also among several operators. The
question that remains is the utility of such measurements on various edges.
In this regard the choice of an optimum definition of the width of an edge de-
serves some study. It appears that the use of such a technique would lead to
some improvement over strictly "eyeball" RES measurements, but the significance
of the improvement is unknown.
Rotating prism scanners have been developed by at least two companies.
These products are discussed below.
Grant Instruments - Spectrum Line Measuring i.