利用三维激光扫描数据和KNNS-ICP算法进行变形分析

利用KNNS-ICP算法对扫描物体的变形量进行分析,提取目标物变形量的6个参数,即3个旋转变量和3个平移变量,充分利用了扫描点云中的几何信息,通过对目标物点云的整体分析来提取变形量,避免了利用单点进行变形分析的偶然性.ICP算法主要从不同目标物点云数据的获取、变形前和变形后点云的整体配准、对全局配准后的每个目标物点云进行局部匹配来获取变形量等3个阶段来进行.通过理想状况下的实验模型及非理想状况下的实验模型进行实例分析,论证了利用该方法进行变形监测的可行性.

Deformation Analysis Using 3D Laser Scanning Data and KNNS-ICP Algorithm

Analysis of deformation on the scanning object using KNNS-ICP algorithm, the 6 parameter extraction of object deformation, i. e. 3 rotational variables and 3 translational variables, making full use of geometric information of the scanning point cloud, according to integral analysis of the point cloud target to extract deformation, avoiding the use of single point of deformation of contingency analysis. Algorithm of ICP mainly excuted 3 stage from different target acquisition, point cloud data of deformation before and after deformation, to obtain the deformation using local matching of every point cloud target after the overall calibration. According to the analysis of the experimental model under the ideal condition and the non ideal conditions, demonstrating the feasibility of using the method of deformation monitoring.