An Improved Active Contour Model Based on C-V Model
Yan Zhang · Journal of Sichuan University of Science & Engineering · 2014
Aiming at the shortcoming of inaccuracy segmentation in non-homogeneous images of C-V( Chan-Vese) model,the energy equation of C-V( Chan-Vese) model is improved.The local gray fitting function is integrated to area term,so that segmentation can balance both global and local information.In addition,the re-initialization can be avoided because the penalty term forces the level set function to be close to a signed distance function.The segmentation experiments for non-homogeneous images demonstrate that the proposed method can get better results than traditional C-V model.