ON THE SEPARATING CAPABILITY OF CELLULAR NEURAL NETWORKS

The cellular neural network is able to perform different image-process ing tasks depending on the template values, i.e. the network parameter s, used. In the case of linear templates the parameter space is divide d into different regions by hyperplanes. Every region is associated wi th a task, such that all points within that region let the cellular ne ural network perform the desired task. In this paper a lower and an up per bound for the number of regions that can be separated with binary input cellular neural networks are given, thus answering the question of how many different-tasks such a cellular neural network can perform .