摘要
提出一种基于局部邻域几何约束的医学图像配准方法,通过判断初始匹配的SIFT特征点间的几何关系,去除误匹配对;在医学图像中提取全部三维灰度特征,进行第二次误匹配对去除操作,匹配精度高、稳定性强、计算速度快.试验结果表明,本文方法在保证计算速度的前提下,能够在一定程度上提高医学图像特征匹配的准确率.
A geometric constraint method based on local neighborhood is proposed to eliminate the wrong matching by judging the geometric relationship between the initial matched SIFT feature points;All the three-dimensional gray features are extracted from the medical image,and the second false matching pair is removed.Finally,the matching result with high accuracy,strong stability and fast calculation speed is obtained.Experimental results show that the proposed method can improve the accuracy of medical image feature matching to a certain extent on the premise of ensuring the computational speed.
作者
王丹
刘太辉
刘丽君
WANG Dan;LIU Taihui;LIU Lijun(Computer Science and Technology College of Beihua University,Jilin 132013,China)
出处
《北华大学学报（自然科学版）》
CAS
2021年第4期556-560,共5页
Journal of Beihua University（Natural Science）
基金
吉林省教育厅科学技术重点项目(JJKH20210049KJ)
吉林省科技发展计划重点项目(20200404223YY).
