关键词: 医学图像分割, 二维Otsu方法, 水平集, Chan-Vese模型

Abstract: This paper presents an OCV medical image segmentation algorithm based on two-dimensional Otsu method and Chan-Vese model.First,we use two-dimensional Otsu method segment the image into the objectives,background,marginalized and noise of four parts then through an energy function to judge edges and noise pixels belonging to the background or target,and the use of regional homogeneity of the global information initial segmentation results tuning,to be more precise segmentation.The algorithm optimizes the initial profile position to address effectively the impact of the initial location of the evolution curve speed,by through point by point to the generation of energy function for the minimal value,to reduce the amount of computation,improve the speed of image segmentation.Experimental results show that the proposed algorithm can resist noise,the result is good,a very good practical significance.

Key words: medical image segmentation, two-dimensional Otsu method, level set, Chan-Vese model