A convex approaches toward global minimization for fast multiphase image segmentation

The study is to investigate a fast multiphase image segmentation model from a statistical framework. The globally convex image segmentation method and the split Bregman method are incorporated into the model by maximizing the posterior image densities over all possible partitions of the image plane. The proposed model is robust with respect to noise based on Gaussian kernel function and can avoid a n unambiguous segmentation due to a new representation of a partition of an image domain into a number of regions. Given these advantages, the proposed method can get good performance and experiments show promising segmentation results on both synthetic and real images.