Design for robustness edgegray detection CNN

Abstract

The cellular neural/nonlinear network (CNN) is a powerful tool for image and video signal processing, robotic and biological vision. The paper sets up a theorem to design a robustness template CNN for edge gray detection in gray-scale, which provides parameter inequalities for determining parameter intervals for implementing the corresponding tasks. The edges in two different kinds of gray-scale image are successfully detected by the edgegray CNNs designed via the theorem.