A novel segmentation model with dual level set function based on Chan-vese and local binary fitting models

Image segmentation is one of the most important operations in the sphere of image processing and pattern identification. At present, most of the research is based on the local segmentation or global segmentation of the image. In this paper we propose a novel segmentation model with dual level set function based on Chan-vese and Local Binary Fitting Models. The structure of the novel model is in view of the means of local and global statistic information. Two evolution curves are defined to realize the local and global segmentation of the image. In order to accelerate the segmentation, the method of additive operator splitting is applied to solve the nonlinear parabolic partial differential equation. The proposed algorithm has be carried over into synthetic and real images. Experimental results show that the novel model can accomplish global segmentation and selective segmentation of the image with intensity in homogeneity, low contrast and weak edge at the same time.