A method for edge detection in gray level images, based on cellular neural networks

Abstract

Edge detection is an important preprocessing task in artificial vision systems. In this paper the utility of a recently reported CNN template for edge detection was verified over a set of black and white images. These images were obtained applying an threshold procedure to their corresponding associated gray level images. An optimal threshold value for preserving a large number of features from the original gray level input images was used. Combining the threshold and edge detection templates, a procedure to obtain edges on gray level images was implemented.