摘要
本文提出一种基于图切割结合区域生长算法的超声图像分割方法.文中采用对图像进行分块后进行图切割的方法,大大降低了算法的时间复杂度,在分割的过程中利用先验知识和图像边缘信息,对分割进行约束和指导,并与区域生长算法相结合从而实现了超声图像的快速自动分割.实验结果证明该方法分割效果良好,具有一定的应用和研究价值.
On this paper, A method is presented for the medical Ultrasonic Image Segmentation which is based graph cuts combined with the region growing segmentation algorithm. In order to reduce the time complexity, Graph cuts is used for the image that is cut to some blocks. And on the segmentation process, it use priori knowledge and edge information to guide the split. Combined with the region growing algorithm, it achieve the rapid and automatic segmentation of medical Ultrasonic Images. The experimental results show that the proposed approach is favorable.
出处
《中央民族大学学报（自然科学版）》
2012年第4期35-39,共5页
Journal of Minzu University of China(Natural Sciences Edition)
基金
福建省教育厅科技项目(No.JA11296)闽教科[2011]08号JA11296
"欧盟科技发展第七框架计划项目-WIDTH"的资助
关键词
超声图像
分割
图切割
区域生长
medical Ultrasonic Image
Segmentation
Graph cuts
region growing
