An ezperiment of medical image segmentation based on Ncut
This paper presents experiments of an approach based on Ncut to recognize object in brain MR images which are relevant with each other. The approach can consider the features in the 3D space through reconstruction weight matrix while using Normalized Cut. The eigenvectors of the weight matrix show segmentation of the images, and the object is recognized. In contrast with the original, the improved method decrease the running time. The mehtod proposed in this paper can segment a series brain MRI images in one time, and the result of the experiment is satisfactory.
Medical Image Segmentation Graph Theory Normalize Cut
Yanzhi Chen Yongfeng Huang
School of Computer Science and Technology,DongHua University,Shanghai,China
国际会议
北京
英文
1-4
2009-06-11（万方平台首次上网日期，不代表论文的发表时间）