基于移动最小二乘曲面的点云拼接算法

Point cloud registration algorithm based on moving least-squares surface

针对现有点云拼接方法受被测零件复杂外形和测量噪声影响的问题,提出基于移动最小二乘(Moving Least-Squares,MLS)曲面的点云拼接算法.该算法利用一种极值投影方法来有效计算点到MLS曲面的对应点,并结合迭代最近点(Iterative Closest Point,ICP)算法将各个视角的点云统一到该MLS曲面.实验结果表明该算法能获得精确、稳定的拼接结果.

Due to the fact that the existing algorithms of point cloud registration are affected by the sophisticated shape and measurement noise of components,a point cloud registration algorithm based on Moving Least-Squares（MLS） surface is proposed.Each point cloud in different view-point are registered onto MLS surface by using a peak value projection method which can obtain the corresponding point on MLS surface effectively and Iterative Closest Point（ICP） algorithm.The experimental results show that accurate and stable registration can be obtained with the algorithm.