【摘要】 点云滤波是机载激光雷达(Li DAR)数据处理中的关键环节。该文就数学形态学滤波方法在坡度参数与高差阈值选取上的不足,提出了一种自适应数学形态学滤波方法。该方法首先由点云建立规则格网并使用当前区域坡度的平均值预测地形坡度参数,然后根据高差阈值与点云密度间的关系对高差阈值的选取进行优化改进,从而进行渐进式形态学滤波得到最终地面点。采用国际摄影测量与遥感学会(ISPRS)所提供的典型地形点云数据进行实验验证,结果显示该文改进滤波算法能更好的区分地面点和地物点,并且能有效降低总误差。更多还原

【Abstract】 The point cloud filtering is the key link in the data processing of airborne Li DAR. In this paper,based on the shortage of mathematical morphology filtering method in selecting the slope parameters and height difference threshold,an adaptive mathematical morphology filtering method is proposed. This method first builds a regular grid from point cloud,the topography slope parameters is predicted by the average value of the current area slope. Then,based on the relationship between the height difference threshold and point cloud density,the selecting of height difference threshold is optimized and improved.Then,the final ground point is obtained by progressive morphological filtering. It is verified by the point cloud data of typical terrain which is provided by the International Society for Photogrammetry and Remote Sensing( ISPRS). The results shows that the improved filtering method can distinguish the ground point and feature point better,and it can effectively reduce the total error.更多还原