MRI脑部图像头骨剥离方法研究

头骨剥离是脑组织分割过程中的关键环节.针对传统的MRI脑组织分割中头骨剥离速度较慢,分割精度不高的问题,提出一种基于水平集的头骨剥离方法.首先根据脑部图像灰度特征确定初始分割曲线,然后将曲线看成高维函数的零水平集,最后利用灰度和形态特征控制高维函数水平集演化,通过高维函数的演化确定曲线的演化,使曲线达到待分割图像边缘.对基于水平集的方法和现有的典型方法进行仿真,仿真结果表明该方法在分割精度上有明显提高.实验证明方法可精确地将脑部组织和非脑组织分离,达到头骨剥离的目的.

Skull-stripping MR Brain Images

Skull-stripping has become a key step in brain tissue segmentation.The traditional methods for MRI image skull-stripping are very slow and not precise.Aiming at this problem,the paper develops a method based on level set.Firstly,it defines the initial segmenting curve based on the gray characteristic of the image.Secondly,it judges the curve as the zero level set of the high-level function.Lastly,it evolves the high-level function by gray and morphological characteristics,then recognizes the edge of MRI bra...