作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2012, Vol. 38 ›› Issue (16): 207-211. doi: 10.3969/j.issn.1000-3428.2012.16.054

• 图形图像处理 • 上一篇    下一篇

基于加权互信息的多模图像配准算法

张峻豪,孙 焱,詹维伟   

  1. (上海交通大学软件学院,上海 200240)
  • 收稿日期:2011-12-06 修回日期:2012-01-10 出版日期:2012-08-20 发布日期:2012-08-17
  • 作者简介:张峻豪(1986-),男,硕士研究生,主研方向:医学图像处理,数字医疗;孙 焱,讲师;詹维伟,教授

Multi-modality Image Registration Algorithm Based on Weighted Mutual Information

ZHANG Jun-hao, SUN Yan, ZHAN Wei-wei   

  1. (School of Software, Shanghai Jiaotong University, Shanghai 200240, China)
  • Received:2011-12-06 Revised:2012-01-10 Online:2012-08-20 Published:2012-08-17

摘要: 基于互信息方法的医学配准容易出现局部极值现象,导致准确率下降。为此,提出一种基于加权互信息的多模图像配准算法。通过全局滤波和边缘提取进行图像预处理,突出图像特征,采用加权互信息方法实现配准。实验结果表明,该算法能够提高多模图像配准的准确率,加快匹配速度。

关键词: 医学图像配准, 局部体积插值, 局部极值, 加权互信息, 预处理

Abstract: In medical image registration based on Mutual Information(MI) method, it exists the problem about the local extrema. Aiming at this problem, this paper proposes a multi-modality image registration algorithm based on Weighted Mutual Information(WMI). Through the preprocessing procedure by global filtering and edge extraction, it extrudes the image features, and uses WMI method to realize registration. Experimental result shows that this algorithm can improve the accuracy and the speed in mulit-modality image registration.

Key words: medical image registration, local volume interpolation, local extrema, Weighted Mutual Information(WMI), preprocessing

中图分类号: