Research On Image Processing Based On Pulse Coupled Neural Network

With the continuous development of information,the popularity of computers and the progress of multimedia technology image processing technology has been applied to all aspects of human life.Image enhancement and segmentation are the important theoretical basis of image processing and it's necessary for us to make a thorough study of it.In recent years,more and more people combine the image processing technology with the human visual mechanism,and put forward some mathematical models with visual characteristics.The pulse coupled neural networks(PCNN)is a mathematical model for simulating the conduction characteristics of the mammalian visual nerve.It has been widely used in the field of image enhancement and image segmentation because it has the characteristics of nonlinear modulation,variable threshold,interaction between neurons,integrated space-time characteristics,and synchronous distribution of similar clusters.In this paper,the basic theory of image enhancement and image segmentation is described first,and The evaluation criteria and common methods of image enhancement and segmentation are introduced,then the advantages and disadvantages of these methods are summarized,so that the follow-up work will be carried out smoothly.Secondly,the network model of the pulse coupled neural networks(PCNN)is introduced,and other principles of work are expounded through its model,then the main characteristics of the pulse coupled neural network and the effects of these characteristics on the pixels in the image are analyzed.Then the basic application of pulse coupled neural networks in image enhancement is studied.The Contourlet transform,which is improved on the basis of the wavelet transform,is combined with the PCNN to enhance the image.I proposean Image enhancement algorithm based on Directionlet transform and PCNN(D-PCNN).An improved method of some parameters in D-PCNN is put forward,which makes it more consistent with the image characteristics to get better image enhancement results.In the end,the application of pulse coupled neural network in image segmentation is studied.The image is processed by the parameter improved PCNN model,and the segmentation results of the image are obtained.At the same time,this paper introduces the two-dimensional entropy with spatial characteristics as the quality evaluation standard for the segmentation effect.On this basis,an adaptive segmentation algorithm is proposed to prevent excessive redundant iterations.