Edge Detection Based On Nearest Neighbor Linear Cellular Automata Rules and Fuzzy Rule Based System

Edge Detection is an important task for sharpening the boundary of images to detect the region of interest. This paper applies a linear cellular automata rules and a Mamdani Fuzzy inference model for edge detection in both monochromatic and the RGB images. In the uniform cellular automata a transition matrix has been developed for edge detection. The Results have been compared to the other classic methods for edge detection like Canny, Sobel, Prewitt and Robert. For performance evaluation, and comparison with the other methods the MSE, PSNR (Peak Signal-to-Noise Ratio), SNR(Signal-to-Noise Ratio) criteria have been used. The Comparison results reveals the superiority of the proposed methods in this paper compared to the other standard edge detection methods.