文章摘要
陈伟卿,欧宗瑛,李冠华,韩军,赵德伟,王卫明.基于互信息与梯度相似性相结合的医学图像配准方法[J].,2009,(3):387-390
基于互信息与梯度相似性相结合的医学图像配准方法
Medical image registration based on mutual information combined with gradient similarity
  
DOI:10.7511/dllgxb200903015
中文关键词: 图像配准  互信息  梯度相似性
英文关键词: image registration  mutual information  gradient similarity
基金项目:“八六三”国家高技术研究发展计划资助项目863-306-ZD13-03-6;大连市科技局科技计划资助项目2005E21SF134.
作者单位
陈伟卿,欧宗瑛,李冠华,韩军,赵德伟,王卫明  
摘要点击次数: 956
全文下载次数: 568
中文摘要:
      针对传统互信息配准方法未利用图像空间信息的缺点,提出一种将互信息与梯度相似性结合的医学图像配准方法.待配准图像的每组对应点的梯度相似性包括方向相似性和模值相似性.待配准图像整体梯度相似性系数由各对应点对的梯度相似性之和决定,该系数与传统互信息的乘积作为图像配准的测度.利用2D多模图像分别进行平移、旋转、采样,得到配准函数曲线,并给出具体的配准实例.实验结果表明,该方法比传统互信息有更高的鲁棒性和精度.
英文摘要:
      To solve the drawback that typical mutual information-based registration neglects the spatial information of images, a new medical image registration method is developed by combining mutual information with gradient similarity. The gradient similarity of each pair of corresponding points includes direction similarity and module similarity. The summation of the similarity term for all sample pairs gives the gradient similarity of two registration images, which is multiplied by the mutual information to form the final registration metric. Registration functions are analyzed and compared, applying in different transform such as translation, rotation and sub-sampling of 2D multi-modal images, respectively. Experimental results show that the new method performs better than typical mutual information in robustness and precision.
查看全文   查看/发表评论  下载PDF阅读器
关闭