transform contains all of the information contained in the original

image. However, the information is now arranged according to spatial

frequency rather than spatial location. The advantage of such an

arrangement is that objects or signals of interest may overlap with

noise in the image domain but exist isolated in the frequency domain.

This permits the possible separation of signal from noise in the

frequency plane when it would have been impossible in the image plane.

The image can be transformed into frequency space, frequency filtered

and then transformed back into image space with the noise removed.

The frequency filter may be low-pass, high-pass, or band-pass, chosen

to optimize the filtering of a specific signal. This frequency plane

filter is the heart of the analog optical computer.

 The frequency plane filter can be constructed in many ways. Low-

pass and high-pass filters are accomplished using simple apertures

mounted on axis in the frequency plane. More complicated filters are

produced optically using holographic techniques. These filters may

also be produced using computer-generated holography (CGH). The

computer is used to model the desired filter response, mathematically

represent the holographic filter, and create a physical filter using a

writing device. One of the important advantages of computer-generated

holography is that the reference need not exist physically, but only

mathematically. This permits mathematical manipulation of the

reference prior to creation of the filter for purposes of

optimization.

 The advantage of an analog optical processor is that it may

operate at very high speeds. In addition, the processor typically is

smaller, lighter, and consumes considerably less power than an