Edge Detection and Image Segmentation Based on Cellular Neural Network

On the basis of brief description of cellular neural network, the process of edge detection based on CNN is introduced with the flow chart of whole algorithm designed, and several kernel techniques are explained respectively in details. As far as binary and gray images, the two simulation models for image edge detection based on CNN and traditional arithmetic operators (prewitt, sobel, canny) respectively are designed and compared their performance. Experimental results demonstrate that the CNN algorithm has several advantages, such as high speed parallel calculation on hardware, calculation speed independent of image size, real-time performance and so on. Therefore, CNN is an effective method for edge detection and image segmentation.