Image segmentation method based on improved fuzzy Chan-Vese model

Medical image segmentation is a hot topic in the field medical image processing. The segmentation methods based on level set and the ones based on fuzzy set are currently very popular in the field of medical image segmentation. But these methods do not balance between global and local features of the image. This paper combines the advantages of these two methods, proposes a fuzzy Chan-Vese model, which introduces fuzzy clustering into Chan-Vese model. This model extends the regional energy part of Chan-Vese model to regional energy based on fuzzy clustering, meanwhile adds fuzzy cluster objects as the constraint of the model, so it can take account of global and local features of the image. In the medical image segmentation experiments, this paper uses OTSU method to execute initial segmentation for getting the initial segmentation curve, and then uses fuzzy Chan-Vese model to realize image segmentation. Experimental results show that, with the help of prior knowledge of segmentation prototypes of medical images, the proposed method has achieved very good segmentation results.