Image Segmentation And Image De-noising Based On Improved PCNN

Image de-noising and image segmentation have laid the foundation for the image recognition and analysis,but also are key steps in computer science.As a basic technology of image preprocessing,image de-noising and image segmentation have always been hot issues concerned by scholars.Pulse coupled neural network(PCNN)model is one of the emerging artificial intelligence neural networks,which is widely used in image processing,analysis,and computer vision.In this paper,we take the PCNN model as the research framework,and propose image segmentation technology based on improved PCNN model and image de-noising technology based on improved bilateral filtering model and PCNN model.The main contents of this paper are the following aspects.(1)A simplified PCNN model is proposed.Parameters of the basic PCNN model are researched and analyzed.Through the analysis on it,combining with the information contained in the image itself,the simplified PCNN model is given.(2)The improved PCNN model is put forward,which combines with spiral optimization algorithm.Parameters are manually selected by doing a lot of experiments repeatedly on some key parameters in image segmentation.(3)A new image de-noising algorithm is raised,which combines bilateral filtering and the PCNN model.Impulse noise is removed through discussing the PCNN model and analyzing its advantages and deficiencies for other noise in image de-noising.Bilateral filtering model is introduced in this paper,to compensate for the lack PCNN model when de-noising.Based on the PCNN model and bilateral filtering model,improved image de-noising technique is proposed.Compared with other methods,the effectiveness of this algorithm is proved in image processing.