激光雷达点云数据截面提取计算方法研究

摘要

The scattered point cloud obtained by LiDAR cannot be used in practical applications. Considering this situation, the paper studies the visualization of LiDAR point cloud and the method to quickly intercept approximate section by the point cloud data interactive calculation. The experiment adopts octree traverse for the point cloud selection, generates approximate cross-section by the box selected polygon, and then achieves vertex fetch and measurement. Through the experiment on an area point cloud data, the cross-section generated can better reflect the terrain changes, and even shows the vegetation distribution patterns in the cross-section. By establishing the index relationship of the 3D point cloud, the method can be used to reduce the interactive calculation amount, to improve the efficiency of the point cloud interaction. It can be applied to process a mass 3D data, topographic surveying, power line design and so on.%　　通过激光雷达获得的散乱点云，通常无法直接应用于实际工作。文中研究了激光雷达点云的可视化以及点云数据交互式计算快速截取近似断面的方法，利用八叉树遍历进行点云选择，通过框选多边形生成近似横截面，并进行顶点拾取和测量。通过对某地点云数据实验，得到的横截面可以较好地反映地形变化，甚至可以显示出植被在该截面上的分布形态。文中方法通过建立三维点云的索引关系，减少了交互中的计算量，提高了点云交互的效率，适用于大数据量三维点云的处理，可应用于地形测绘、电力勘察设计等领域。