A Novel Automatic Parameter Setting Method of PCNN for Image Segmentation

Image segmentation plays an important role in the image processing and the pulse coupled neural network (PCNN) has good performance in image processing. It is the parameters of PCNN that hinder the application of PCNN, which are usually determined through trial and error. This paper presented a novel method to determine the parameters of the standard PCNN model and applied it to image segmentation, using gravitational search algorithm (GSA). In the proposed algorithm, the information entropy of the image is used as the fitness function of GSA and Peak Signal to Noise Ratio (PSNR) is used as an evaluation indicator of the results of the experiment. Compared with the state-of-the-art algorithm, such as Otsu, Kmeans, Genetic algorithm and manual adjustment method, the experimental results showed that the proposed algorithm GSA-PCNN is more efficient and accurate.