Research On Medical Image Segmentation Based On Fuzzy C-Means

As an important carrier of information,medical image plays a more and more important role in clinical diagnosis.Among them,segmentation of magnetic resonance images in medical images can help doctors quickly understand diseases and other information,which has been widely used in various kinds of medical diagnosis and research.In medical image segmentation algorithms,fuzzy C-means(FCM)clustering algorithm has become one of the mainstream segmentation algorithms due to its easy execution and fast running speed.However,FCM is sensitive to the initial clustering center and noise,so it is difficult to effectively balance the relationship between noise reduction and detail retention,and improper processing will directly affect the segmentation performance.As for these problems,this paper mainly does two aspects of work.On the one hand,in order to reduce the noise and select appropriate clustering centers,an improved fuzzy C-means model fused with particle swarm optimization is proposed.The model uses the grayscale and spatial information of image pixels to construct a new weighting method of image grayscale value,and performs morphological operations on the image to obtain a new grayscale processed image.At the same time,the inertia weight of particle swarm optimization algorithm is improved to replace the iterative updating method of FCM clustering center,so that it can get adaptive updating according to the image segmentation situation.Experiments on different medical images show that the proposed algorithm can effectively reduce the noise and the influence of clustering center,and improve the segmentation performance of medical images.On the other hand,a new FLICM algorithm is proposed by modifying its neighborhood item.Firstly,the original image is processed by non-local mean filtering to generate additional images,and the pixel consistency coefficient is defined according to the additional image and the original image.Then,the pixel difference coefficient between the center pixel and the neighborhood pixel is defined by the image local information,and the pixel correlation coefficient is constructed by combining the pixel consistency coefficient with difference coefficient.Finally,the fuzzy factor of FLICM is replaced by the pixel correlation coefficient,and the improved FLICM algorithm is obtained.Experiments on different images show that the proposed algorithm can obtain more accurate segmentation results.