| 海淼,沈毅.三维多模态医学图像自动配准方法研究[J].中国医学影像技术,2005,21(5):798~801 |
| 三维多模态医学图像自动配准方法研究 |
| Research on automatic registration method for 3D multi-modal medical images |
| 投稿时间：2004-12-08 修订日期：2005-01-21 |
| DOI： |
| 中文关键词: 医学图像配准 图像分割 互信息 数学形态学 |
| 英文关键词:Medical images registration Image segmentation Mutual information Mathematical morphology |
| 基金项目: |
|
| 摘要点击次数: 1996 |
| 全文下载次数: 980 |
| 中文摘要: |
| 提出一种综合应用图像分割与互信息的医学图像自动配准方法。首先采用门限法和数学形态学方法进行预处理,再用k-means方法进行分割,之后采用基于互信息的Powell优化方法配准。将该方法用于磁共振图像(MRI)和正电子发射断层扫描(PET)临床医学图像配准,得到较满意的效果。 |
| 英文摘要: |
| An automatic algorithm was presented based on image segmentation and mutual information for multi-modal medical images. Images were firstly preprocessed using threshold method and mathematical morphology, secondly segmented using k-means algorithm, and then registered using powell method based on mutual information. The experimental Results of clinical multi-modal images showed that the proposed medical images registration method works well and even faster than most of existing registration methods. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
