图像多相分割松弛凸化模型分裂方法

Split Method for the Convex Relaxation Model of Image Multi-Phase Segmentation

研究了一类向量值极小化问题的凸松弛方法,给出了适用于split Bregman快速算法的一般性等价模型。Vese-Chan多相分割方法和基于分片常数水平集函数的Mumford-Shah方法是新模型的特例。数值实验表明,在Vese-Chan方法和Mumford-Shah方法中应用split-Bregman算法,具有较快的运算速度和较好的分割效果,且对初始条件是鲁棒的。

A general equivalent model is introduced based on the convex relaxation model of a class of vector-valued minimization problems. The presented model can be solved by split-Bregman algorithm. The computational efficiency is greatly improved. The method is applied to the Vese-Chan multi-phase segmentation model and Mumford-Shah model. Numerical experiments show our method has fast computing speed and good segmentation results, and is robust to the initial condition.