Method of edge detection based on non-linear cellular automata

The quality of edge detection is usually unsatisfactory for real-world image with many textures. This paper introduces a new method for edge detection based on non-linear cellular automata. Compared with other four classical methods, such as the Robert, Sobel, Prewitt, and Log detectors, the experimental results have shown that this method can not only remove the noise effectively, but also ensures the continuity, integrity and accurate location of imagepsilas edge.