基于最大类间方差法的脑MRI图像分割

摘要

In specific brain MRI image segmentation,classic level set method has the disadvantages of time consuming,segmentation slow and not be able to convergence to the actual edge of the object.We adopt the method of the Maximum Between-cluster（Ostu）.The paper states the application to the specific brain MR_I segmentation in detail and compares the results of edge detection between Ostu and the level set method.The experimental results showed that Maximum Between-cluster not only simple,fast,but also can detect the edge accurately.%针对具体的脑MRI图像分割，经典的水平集方法存在处理时间较长、分割速度慢、不能很好地收敛到物体的实际边缘等缺点，本文采用了最大类间方差法（Ostu法）。文章详细阐述了最大类间方差法在具体的脑MKI图像分割中的应用，并将基于该方法的脑MRI图像边缘检测的实验结果与经典的水平集分割方法的处理结果进行了比较。实验结果显示，最大类间方差法不仅原理简单、运行速度快，而且能够实现边缘的精确检测。