基于改进的Mean Shift算法虚拟人脑图像分割

Digital Human Brain Image Segmentation Based on an Adapted Mean Shift Method

为了克服Mean Shift算法各向同性的缺点,使用结构信息构造各向异性高斯核,使其具有各向异性,从而克服细长目标的影响;将颜色空间投影到新的坐标系下,使得相近颜色可以有较大的距离,以增大虚拟人脑图像中灰质与下层数据之间的区别.虚拟人脑图像分割结果说明,该算法可以得到较好的分割结果.

In order to overcome the limitation of the Mean Shift method, this paper presents a new anisotropic Gauss kernel, based on structure information, and by the Gauss kernel newly proposed, the new model can reduce the effect of gracile topological structure. In addition, we project the color space to a new space, based on PCA model, to expand the distance of similar color and enlarge the difference between grey matters and grey matters belonging to next picture. The results of the segmentation of the digital brain image show that better results could be achieved by the adapted Mean Shift method.