摘要
Abstract
In view of the point cloud filtering problem in airborne light detection and ranging(LiDAR)data processing,this paper selected four commonly used filtering methods:simple morphological filtering(SMRF),cloth simulation filter(CSF),slope-based filter(SBF),and progressive triangulated irregular network densification(PTD).The paper also conducted a comparative study on the four methods in four typical scenarios(industrial areas,urban areas,artificial forests,and natural forests).The study evaluated the accuracy and suitability of the four filtering methods in different scenarios through a combination of quantitative analysis and qualitative analysis.The results show that the SMRF method has the highest accuracy and stability in most scenarios,especially in the removal of buildings and terrain preservation.The CSF method and SBF method have different performances in different scenarios,and the PTD method performs well in the natural forest scenario but has larger errors in densely built-up areas.关键词
机载激光雷达(LiDAR)/点云滤波/多场景/精度评价Key words
airborne light detection and ranging(LiDAR)/point cloud filtering/multi-scenario/accuracy evaluation分类
天文与地球科学
