Abstract

It is generally acknowledged that the performance of an optical correlator is dependent on the input scene contrast. The magnitude of this dependence has yet to be determined, however. As optical correlators begin to move from the laboratory into less benign environments, a better understanding of this effect is required. This presentation will focus on two aspects of the problem. First, given a filter made from a perfect, i.e., high contrast, target, how do changes in the input scene contrast affect the correlation peak to noise ratio? Second, given that the input scene will be low contrast, should the filter be made from a high contrast target or a low contrast target? The results of experiments designed to answer these two questions for a limited number of targets and backgrounds will be given and some general conclusions drawn.

© 1992 Optical Society of America