抗差趋势面的机载LiDAR点云滤波方法

传统曲面约束滤波算法中,利用最小二乘拟合地形曲面易受种子点粗差影响。针对这一问题,提出基于抗差趋势面的机载激光雷达点云数据滤波方法,首先构建格网索引组织数据,引入抗差趋势面拟合合理的区块地形,通过自适应阈值的设置实现不同区域的自动灵活处理,最终滤除孤立点完善滤波结果。使用ISPRS提供的测区数据进行实验,与传统曲面拟合方法进行对比,实验结果证明,该方法较传统移动曲面拟合法能够得到更加可靠的滤波结果,具备较高实用价值。

Airborne LiDAR Point Cloud Filtering Method Based on Robust Trend Surface

Least Squares Adjustment is used to fit block terrain in traditional moving curved fitting filtering method and the disadvantage of this method is sensitive to outliers. Aiming to solve this disadvantage, the method was pro-posed for filtering LiDAR point cloud based on robust trend surface in this paper. Firstly, the grid index was con-structed to organize LiDAR point cloud data, robust trend surface was introduced to fit block terrain reasonably, then self-adaption threshold was set to deal with different area more flexibly, and isolated points were removed fi-nally to perfect the filtering result. The test data provided by ISPRS were adopted for experiment; traditional curved fitting filtering method was selected for comparison. Experimental results proved that the proposed method is practi-cal as the filtering results are more reliable than traditional moving curved fitting filtering method.