Grayscale Image Edge Detection Based on Pulse-Coupled Neural Network
Dong Sheng Ji · Jisuanji gongcheng · 2003
A novel approach is proposed to calculate the edge value of the central pixel of a small window. When applying the edge values to modulate the pulse of the corresponding neurons in a PulseCoupled Neural Network (PCNN), one can reduce the noise and obtain a precise edge of a grayscale image based on the property of synchronization of the network. Furthermore, an image enhancement method, which increases the contrast of the image using the output of the network, is used to improve the edge detection in unsharp regions. The experiment results and comparisons show that this method is effective and feasible.