A projection model for Chan-Vese model in level set formulation and its Split Bregman algorithm
Zhimei Zhang, Zhenkuan Pan, Guodong Wang, Ximei Zhao, Jieyu Ding · 2011
In this paper, we propose a variational projection model to diffuse the evolution of Chan-Vese model (i.e., active contours without edges) to be close to a signed distance function, for which has a good geometric feature in level set function formulation. Then, by introducing an auxiliary variable and a Bregman iterative parameter, we build a fast numerical algorithm-a Split Bregman algorithm, for the proposed projection model. Finally, we apply the algorithm to both simulated and real images with different character and get the effective results.