selected, the transmission is set to one but the phase assumes values
of zero or pi.
The pre-processing and modulation options are the same as in the
previous cases but there is an additional choice in the type of binary
hologram. As was discussed in Chapter III, many mappings are possible
to convert the complex filter function to a real binary pattern.
Presently, Lohmann, Lee, and Allebach-Keegan (A-K) type holograms are
available to the CGH algorithm and the simulation. Figures 7.15,
7.16, and 7.17 show the A-K hologram of the square using no pre-
processing, frequency emphasis and phase-only filtering. Figures 7.18
and 7.19 show the auto-correlation of the square using the A-K hologram
with frequency emphasis and phase-only filtering. Table 7.3 shows the
signal-to-noise and the efficiency for the auto-correlation of the
square using the A-K hologram.
An Example Using an SDF as a Reference
The auto-correlation of the square is theoretically interesting
and provides a common tool by which various techniques can be
compared. Additionally, the auto-correlation of a square is a problem
which can be solved analytically for many of the types of holograms,
lending credibility to the simulation results. However, the real
power of the simulations occur when they are applied to more
complicated imagery. Actual images with complicated shapes and
patterns are impossible to correlate analytically but rather must be
correlated by a computer. Figure 5.1 shows various images from a
training set which was used to create the Synthetic Discriminant
Function (SDF) shown in Figure 5.2. The SDF was created from 36 views
of the object rotated 100 between each view. The SDF should therefore