激光扫描点云数据预处理技术研究  

Pre-processing technology based on point cloud from laser scanning system

作　　者：郭迎福[1] 王晓强[1] 李兵[1] 陈安华[2] 李鹏南[2] 

机构地区：[1]西安交通大学精密工程研究所,陕西西安710049 [2]湖南科技大学机械设备健康维护省重点实验室,湖南湘潭411201

出　　处：《西安科技大学学报》2006年第3期408-411,415,共5页Journal of Xi’an University of Science and Technology

基　　金：国家自然科学基金资助项目(50275120);湖南省杰出中青年基金项目(02JJYB009);湖南科技大学机械设备健康维护省重点实验室开发基金资助项目(KFJJ0503)

摘　　要：激光扫描系统得到的点云数据,密度大,包含有大量的冗余数据。针对这种数据的特点,对切片上的点云数据,提出了给定误差下的最小二乘非均匀B样条拟合法压缩点云数据的方法;对切片数目的精简,提出了累加弦长,并等分弦长,然后反算等分点在总弦长的位置来精简切片数目的方法。通过实例验证表明,该方法对点云数据处理的灵活性和适应性都较好,能够满足曲线和曲面重构的要求。The point cloud from laser scanning system has a high density, and lots of redundancy data are included. Based on this characteristic, an algorithm of least - square non - uniform B - Spline under the given error was put forward to compress the section point cloud, also an algorithm of chord accumulating, chord division and calculating the position of division points on the chord to compress the amount of section point cloud was given out. An example is given to testify that the method is effectual, feasible and meet the require of the reconstruction of curves and surfaces.

关 键 词：反求工程 曲面重构 数据预处理 激光扫描 

分 类 号：TP391.72[自动化与计算机技术—计算机应用技术]