基于最大类间方差法的肝脏CT图像分割及实现

针对肝脏CT图像中,肝脏组织与周围组织之间密度差别不大而易造成漏分或过分的现象,实现了一种能得到最优分割阈值的方法,即最大类间方差法.实验结果表明,该算法计算简单,将图像二值化同时得到最优分割阈值,从而能较好地分割出肝脏组织.文中也对该算法存在的问题进行了分析讨论.

Realization of liver CT image segmentation based on Otsu algorithm

For the little density difference between the liver tissue and the surrounding tissue that easily cause leakage or excessive segmentation. This paper implements an optimal segmentation threshold method,namely Otsu. The experimental results show that the method is simple, can get the optimal segmentation threshold so that it can better segment the liver tissue. The article also analyzes the problems of this algorithm.