A simple postprocessing method for an optical correlation based real-time human face recognition processor.

The use of similarity measure techniques on the optical correlation output plane is analysed as a method to improve the discrimination ability of pattern recognition procedures. The study is divided into two parts: one recounts a computer simulated example corresponding to pattern recognition by the use of input images that may be blurred, rotated or corrupted by additive Gaussian noise, and the second part describes a real optical processor for real-time human face recognition, in which the simple similarity measure based postprocessing method plays an important role for improving the discrimination ability while saving much processing time. [ABSTRACT FROM AUTHOR]