摘要
根据核医学图像的特点,提出一种基于图像轮廓和灰度互信息的核医学图像配准算法.该算法首先基于图像的轮廓的基本特征参数确定粗配准的尺度参数、平移参数和旋转参数,再基于粗准结果利用最大互信息配准准则确定精配准的配准参数.对实际核医学图像的配准实验表明,该算法配准精度高、鲁棒性好,同时配准速度比较快.
According to the characteristic of nuclear medicine image,a method of nuclear medicine image registration based on the contour and intensity mutual information are proposed. It determines the coarse matching parameters：the scale parameteg the translation parameters and the rotation parameter based on flae contourS basic charac- teristic parameters of the images, and then it determines the fine matching parameters by using registration criterion of the maximum mutual information based on the coarse registration result-Experiments results on nuclear medicine images demonstrate that this method is a robust and fast registration method with hi 典 accuracy.
出处
《湖南工程学院学报（自然科学版）》
2008年第1期51-54,共4页
Journal of Hunan Institute of Engineering(Natural Science Edition)
基金
广西教育厅科研基金资助项目(200508107)
关键词
图像配准
核医学图像
轮廓
互信息
image registration
nuclear, medicine image
contour
mutual information
