基于Wasserstein距离和分裂Bregman方法的图像分割算法

摘要

Active contours segmentation algorithms based on C-V model can not be used for intensity non-homogeneity images segmentation. Wasserstein distance is employed as the region histogram similarity measure. A non-parametric active contour model based on this measure is proposed in this paper. Global convex segmentation and split Bregman method is applied to solving the model. It can reduce the amount of computation. Experimental results show that the model does not depend on initial curve. It can segment intensity non-homogeneity images correctly and efficiently.%针对基于C-V模型的活动轮廓分割算法无法应用于灰度非均匀图像分割的问题.采用Wasserstein距离作为区域直方图相似性测度,提出基于该测度的非参数活动轮廓分割模型.在模型求解时引入全局凸分割和分裂Bregman方法,减少了计算量.大量实验结果表明该模型不依赖初始轮廓曲线的位置,能够对灰度非均匀图像进行较准确的分割,具有较快的运算速度.