A variational level set method for multiphase image segmentation

Abstract

A novel multiphase image segmentation model under variational level set method framework is presented in this paper. The energy functional includes three parts: the first one is region-based component resulted from maximum a posterior probability of image partition, the second one is based on minimum geodesic active contours, the third one comes from the enforcement of level set functions as signed distance functions to avoid re-initialization. New characteristic functions for 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">N</sup> regions are designed based on N level set functions and modified Heaviside functions. A semi-implicit finite difference scheme is presented for numerical solution of the corresponding evolution equations of level set functions. Some multiphase image segmentation examples are given to validate the model presented in this paper.