摘要
文章针对非刚性配准CPD(coherent point drift)算法的过程中,计算复杂运行时间长的问题,提出了一种基于曲率特征点检测的配准算法。通过提取目标点集的特征点集,并对所提取的特征点集用CPD算法进行配准,可以使减少运算规模进而减少算法运行时间,提高了配准效率。
In order to solve the problem of complex calculation and long running time in the process of non-rigid registration of CPD（coherent point drift） algorithm, a registration algorithm based on curvature feature point detection is proposed. By extracting the feature points of the target point set and registering the extracted feature points with the CPD algorithm, the operation can be less complex and the algorithm running time can be reduced, thereby improving the registration efficiency.
作者
刘咏璐
Liu Yonglu(1.AnHui Province Key Laboratory of Medical Physics and Technology, Center of Medical Physics and.Technology, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, China; 2.University of Science and Technology of China, Hefei 230026, Chin)
出处
《电子技术（上海）》
2018年第3期14-16,共3页
Electronic Technology
关键词
特征点检测
非刚性点集配准
相干点漂移
feature points extracting
nonrigid algorithm
coherent point drift
