Abstract

A new algorithm and system architecture for inverse filtering using a coherent iterative optical processor is presented. The algorithm theory and convergence is presented along with a computer simulation of the algorithm performance. Two optical architectures for the implementation of this algorithm are also presented. The optical implementation of this algorithm provides a method for the implementation of optical division of constants, functions of one dimension, and functions of two dimensions. Application of this processor to image restoration will be demonstrated.

© 1992 Optical Society of America