Abstract: Medical image segmentation is an important application field of image segmentation,it widespreadly has high noise,artifacts,low contrast,uneven gray,fuzzy boundaries between different between soft tissue lesions and other characteristics,this paper used clustering algorithm,combined with LCM and two phase model level set method (CV),chose the appropriate filter for medical image denoising,then used the fuzzy c-means algorithm to get image prior model.And we improved the traditional CV model to fine the image segmentation.Experiments show that the model can solve the problem of high image noise and weak boundary,and can effectively avoid the re-initialization,and is more sensitive to the edge,improving the segmentation accuracy,suppressing noise effectively,significantly reducing the number of iterations and time,having certain application value.

Key words: Fuzzy c-means clustering,Filter,LCM model,FCM-CMCV level set method