一种基于最小二乘固定球法的多视点云拼合技术

A Technology of Registration and Integration for Multi-view Point Cloud Data based on Least-square fixed Sphere

为了得到复杂实物的完整测量数据,需要将不同视角下测得的点云统一到同一个坐标系下并进行数据融合。提出了一种基于最小二乘固定球法的多视点云拼合算法;采用拟合最小二乘球法获取球心坐标,然后用三点对齐法实现多视数据的坐标变换,最后用加权平均法去除两组点云中的冗余点实现两组数据的融合。利用UG/OPEN工具集将该拼合技术集成到UG中,应用实例证明了该技术的有效性和实用性。

In order to get the entire point data from complex model, it is necessary to locate and merge the differ- ent - view - measured point cloud data in a global coordinate system. An algorithm of registration and integration for multi - view point cloud data based on least - square fixed sphere is proposed, in which the coordinates of sphere center are calculated by fitting least - square sphere, the coordinates of cloud data are transformed by the superposition of three datum points and the unwanted points are removed to integrate two sets of point cloud data by using the weighed average algorithm. Finally, through integrating the technology with UG by UG/OPEN, the validity and practicability of this technology are proved by an example.