Image Segmentation Based on Local Region LBP algorithm

A method for image segmentation was presented which based on LBP(Loopy Belief Propagation) of local region energy minimization. Based on global energy minimization for image segmentation by traditional LBP algorithm, the proposed method used local region energy as the messages of LBP algorithm, and combined global energy minimization with local region energy minimization, so improved the efficiency of messages passing in LBP algorithm. In the experiments, the proposed method is compared with the traditional LBP algorithm. Experiments on natural images showed that the proposed method is able to obtain more accurate segmentation results and also less sensitive to noise than some of the existing models in common use.