基于不同密度LiDAR数据DEM构建研究  

Research on DEM Construction Based on Different Density LiDAR Data

作　　者：尤号田 YOU Hao-tian(College of Geomatics and Geoinformation, Guilin University of Technology, Guilin,Guangxi 541004)

机构地区：[1]桂林理工大学测绘地理信息学院,广西桂林541004

出　　处：《安徽农业科学》2019年第9期4-7,共4页Journal of Anhui Agricultural Sciences

基　　金：广西科技计划项目(2018GXNSFBA281075);桂林理工大学科研启动基金项目(GLUTQD2017094)

摘　　要：为了进一步提升数字高程模型构建效率,同时降低数据获取成本,以机载LiDAR数据为基础,通过对机载LiDAR数据进行处理得到地面点并对地面点进行抽取操作以获得不同密度地面点数据并插值生成DEM,最终获得不同抽取率下DEM数据生成的精度。结果表明,对于城区而言,随着点云密度的下降DEM生成精度RMSE从0.109 m逐渐增大到0.691 m;对于草地而言,RMSE则从0.065 m逐渐增大到1.096 m;对于林地而言,RMSE从0.088 m逐渐增大到2.201 m。对于3种地物类型而言,DEM生成精度均随抽取率的增大而逐渐降低,且不同地物类型RMSE的变化范围不同。To further improve the efficiency of building DEM and reduce the cost of data acquisition, the airborne LiDAR data were used and processed in this paper. And the ground points were extracted and different density ground points data were obtained using exacting method. Then the DEMs were generated with kriging interpolation method. Finally, the accuracy of generated DEM under different extraction rates were obtained. It was showed that for urban area, the RMSE increased from 0.109 m to 0.691 m with the decrease of point cloud density. For grass area, the RMSE increased from 0.065 m to 1.096 m. For forest area, the RMSE increased from 0.088 m to 2.201 m. It was concluded that the accuracy of DEM generation decreased with the increase of extraction rate and range of RMSE varies with different terrain types.

关 键 词：数字高程模型 激光雷达数据 地面点 抽取率 克里金插值算法 

分 类 号：S127[农业科学—农业基础科学] TP79[自动化与计算机技术—检测技术与自动化装置]