Implementation of a cellular neural network over an optical computer

In the last decade many artificial neural network algorithms have been developed of for different purposes. One of the most important algorithms is the cellular neural network (CNN). This algorithm has been used mainly for image processing and pattern recognition because of its parallel nature and it can be implemented on VLSI circuits. There are a few works using optical computers like Hopfield and BAM networks. In this paper an implementation of the algorithm of CNN over an optical computer is presented.