Abstract

Real-time correlation in the spatial domain has been demonstrated previously by using an acousto-optical (AO) correlator. This acousto-optic system performs correlations at high frame rates, which permits the implementation of real-time pattern recognition algorithms that use multiple templates. Previous work has been done with edge-enhanced templates and images, by using binary rather than full gray-scale pixels. The present work shows the ability of the AO correlator to process gray-scale imagery for the purpose of implementing the classical matched filter. Since intensity invariance is an issue with gray-scale imagery, normalized correlation is implemented on the AO correlator. Experimental results obtained by using real gray-scale imagery are presented.

© 1992 Optical Society of America