摘要
点云配准是通过匹配具有重叠部分的数据集,将不同坐标下的三维数据集变换到同一坐标系下,得到旋转变换矩阵和平移向量。通过各种约束条件,建立适当的模型。刚性配准在变换过程中不会发生形变,而非刚性配准需要考虑形变等问题。刚性配准和非刚性配准在约束条件上有所不同,文章从特征、显著性、正则化等约束条件,对刚性和非刚性配准约束条件进行了研究讨论。
Point Cloud registration is to obtain the rotation transformation matrix and the translation vector by matching the data sets with over-lapping parts, and transforming the 3D data sets of different coordinates into the same coordinate system. The rigid registration in the transformation process will not be deformed, nonrigid registration need to consider the deformation and other issues. Various constraint conditions can be used to establish the appropriate model. Rigid registration and nonrigid registration are different in the constraint conditions, the paper dis- cusses through the features, significant and regularization.
出处
《微型机与应用》
2016年第23期12-14,17,共4页
Microcomputer & Its Applications
关键词
配准
刚性
非刚性
形变
约束
registration
rigid
nonrigid
deformation
constraint