Image Segmentation Algorithm Based On PCNN And Its Application

Image segmentation is one of the basic technologies in the field of image processing.It refers to the process of dividing an image into several independent objects with the same pixel attributes by utilizing some characteristics of the image,such as gray level,color,texture,shape and so on.It is essentially a clustering process based on the pixel attributes.According to the gray level,color,texture,shape and other characteristics of the image,and using various mathematical ideas and tools,people use different models to segment the gray level image and color image,and then form an image segmentation method with complex entanglement.All kinds of methods are interwoven with each other.Interaction,mutual promotion,constitute a complex system.This paper mainly discusses the image segmentation algorithm based on Pulse Coupled Neural Network(PCNN).Firstly,starting from the basic theory of image segmentation,this paper briefly describes the research background of image segmentation technology,and expounds the basic concepts,methods,applications and research status of image segmentation at home and abroad.Secondly,based on the principle of image segmentation and comprehensive analysis of PCNN,color image segmentation method and multi-value segmentation based on PCNN are proposed.PCNN-based color image segmentation method: Firstly,the similarity cluster characteristics of PCNN ignition capture are used for image segmentation,and then the maximum Shannon entropy criterion,the maximum gray-scale entropy criterion and the maximum variance ratio criterion are used for segmentation in RGB color space.Finally,the maximum Shannon entropy is used to filter the iterated binary images,and the filtered binary values are superimposed.Image to obtain a gray level image.Research on multi-value segmentation based on PCNN: Solves the shortcomings of binary image,improves the threshold and pulse in traditional model,and effectively refines and classifies high brightness areas,avoids the phenomenon that traditional model is easy to over-segmentation or even wrong segmentation.Finally,a quasi-optimal PCNN model is applied to image segmentation of anterior segment.The algorithm is simulated by using MATLAB,and the experimental results are analyzed.The results show that the segmentation results of this algorithm are superior to other segmentation algorithms in both subjective and objective evaluation indexes...