复杂采空区多次探测的点云数据精简方法

 针对复杂采空区激光探测中存在探测盲区需要进行多次重复探测的问题，研究激光多点扫描的点云数据精简方法。通过多点探测避免了单次探测盲区，加密了数据稀疏区。通过分析激光扫描轨迹线的拓扑关系，归纳了点云数据的分布特点。在对比传统数据精简的基础上，提出了保留采空区几何特征更为有效的点云数据精简方法--边长角度综合判据法，将密集区域的点云数据进行稀释。验证结果表明，通过对比精简前后求得三维模型的体积、精简率等指标，认为该方法保证了边界三维信息的完整性，而且该方法的数据精简率可达15%~25%。为矿山复杂采空区激光扫描三维空间信息精简获取提供了一种新思路，可后续三维建模及应用奠定基础。

Multiple Detection Data’s Simplification in Complicated Goafs

This paper studies data reduction of laser scanning to solve the problems generated by repeating of laser detection for blind areas. Blind areas of single detection were avoided and data of sparse areas were densified by multi-point detection. Distribution characteristics of point cloud data were concluded by topological relation of laser scanning track analysis. A more efficient point cloud data reduction method, the side angle integrated method, retained the geometric characteristics of goafs proposed in contrast to conventional data reduction, and the point cloud data of intensive areas were diluted. Through comparison of the 3D model volume, reduction rate and other indicators before and after reduction, the verification results show that the method ensures the integrity of boundary 3D information and the data reduction rate reaches 15%-25%. This method provides a new idea for laser scanning 3D information reduction in complex goafs, laying a good foundation for the subsequent 3D modeling and application.