摘要
针对点云数据构建数字高程模型(DEM)的精度受点云抽稀和内插算法等因素影响的问题,设计不同点云抽稀方法和不同内插算法对DEM构建精度影响的两组试验,寻求适用于复杂条带状泥石流沟谷地形构建DEM的组合方法.利用空间抽稀、体素分割抽稀、随机抽稀3种方法以96%的抽稀率对原始点云数据进行抽稀,得到的抽稀结果再分别应用自然邻点插值法、反距离权重法(IDW)、克里金法、样条函数法4种内插算法进行内插,共构建出12种组合生成DEM的方法,对每种组合方法构建的DEM进行高程和坡度精度评价分析.结果表明,针对泥石流沟谷地形空间抽稀条件下利用IDW进行DEM内插的结果最为可靠,拟合优越度为0.999540,平均误差为0.7462 m,中误差为1.2629 m.
Aiming at the problem that the accuracy of constructing a digital elevation model(DEM)from point cloud data is affected by such factors as point cloud thinning and interpolation algorithms,we designed two groups of experiments on the impact of different point cloud thinning methods and interpolation algorithms on the accuracy of DEM construction,and sought for a combination method suitable for constructing DEM from a complex strip debris flow gully terrain.Three thinning methods,namely spatial thinning,voxel segmentation thinning and random thinning,were used to dilute the original point cloud data at a dilution rate of 96%.The thinning results were interpolated by four interpolation algorithms,namely natural neighbor interpolation,inverse distance weighted(IDW)and Kriging and Spline.A total of 12 combined generating DEM methods were constructed,and the elevation and slope accuracy of the DEM constructed by each combined method evaluated and analyzed.The experimental results showed that the results of DEM interpolation using IDW under the condition of spatial thinning of debris flow gully terrain were the most reliable.The fitting degree R2=0.999540,mean error is 0.7462 m,root mean squared error is 1.2629 m.
作者
黄志豪
袁希平
甘淑
杨敏
黎昊
HUANG Zhi-hao;YUAN Xi-ping;GAN Shu;YANG Min;LI Hao(School of Land and Resources Engineering,Kunming University of Science and Technology,Kunming 650093,China;Application Engineering Research Center of Spatial Information Surveying and Mapping Technology in Plateau and Mountainous Areas Set by Universities in Yunnan Province,Kunming 650093,China;School of Earth Sciences and Engineering,Key Laboratory of Mountain Real Scene Point Cloud Data Processing and Application for Universities in Yunnan Province,West Yunnan University of Applied Sciences,Dali 671006,Yunnan,China;Guangdong Open Laboratory of Geospatial Information Technology and Application,Guangzhou Institute of Geography,Guangdong Academy of Sciences,Guangzhou 510070,China)
出处
《兰州大学学报（自然科学版）》
CAS
CSCD
北大核心
2023年第4期460-467,共8页
Journal of Lanzhou University(Natural Sciences)
基金
国家自然科学基金项目(41861054)
云南省基础研究专项(202101AU070123)。
关键词
泥石流沟谷
地面激光扫描
点云抽稀
内插方法
数字高程模型
debris flow gully
terrestrial laser scanning
point cloud thinning
interpolation method
digital elevation model
