MR image segmentation based on level set method

In this paper, a level set model combining probabilistic statistics for image segmentation is proposed. Through adding a single-point pixel distribution into the energy function, the step size of each iteration is increased and the efficiency of the algorithm is improved. By adding the membership function of fuzzy clustering and bias field function, this method can effectively segment the image with intensity inhomogeneities. In addition, a new rule item is added to improve the edge segmentation effect of the image. Experiments on MR images of the brain show that the proposed model can provide ideal segmentation results compared with several level set segmentation models.