摘要
针对三维激光扫描时球形标靶受边缘混合像元影响致使球形标靶点云拟合精度降低的问题,提出了一种基于激光入射角定权的球形标靶点云拟合方法。通过分析球形标靶的特点以及入射角与光斑大小的关系,推导入射角对拟合参数精度的影响关系,确定了球形标靶拟合定权方法,并应用于最小二乘法拟合解算过程中。通过对不同距离下的不同类型球形标靶进行数据采集和处理,对比分析了基于入射角定权方法的拟合结果。实验结果表明,该定权方法具有一定的稳定性,在50 m的扫描距离内球形标靶点云拟合参数精度提高了18.68%~85.63%;该方法较最小二乘法能够有效地减小拟合误差,在20 m的扫描范围内,2种不同类型球形标靶点云的拟合半径误差小于1 mm。
Aiming at the problem that the point cloud fitting accuracy of spherical targets reducing due to the influence of edge blending cells on spherical targets during 3D laser scanning,we proposed a spherical target point cloud fitting method based on laser incidence angle weighting.By analyzing the characteristics of spherical targets and the relationship between the angle of incidence and the size of spot,we deduced the influ-ence of incident angle on the accuracy of fitting parameters,determined the weighting method of spherical target fitting,and applied it to the least squares method fitting solution process.Through the data acquisition and processing of different types of targets at different distances,we com-pared and analyzed the fitting results based on the weighting method of incident angle.Experimental results show that the weighting method has certain stability,and the accuracy of spherical target point cloud fitting parameters is improved by 18.68%~85.63%within a scanning distance of 50 m.The method can effectively reduce the fitting error compared with the least squares method,and the fit radius error of two different types of spherical target cloud is less than 1 mm in the scanning range of 20 m.
作者
程宇翔
郑德华
刘存泰
王浩
胡创
李思远
CHENG Yuxiang;ZHENG Dehua;LIU Cuntai;WANG Hao;HU Chuang;LI Siyuan(School of Earth Science and Engineering,Hohai University,Nanjing 210098,China)
出处
《地理空间信息》
2023年第6期12-16,57,共6页
Geospatial Information
基金
2020年国网新源控股(水电)有限公司科技项目(SGXYKJ-2020-079)。
关键词
入射角定权
球形标靶
点云拟合
最小二乘法
incident angle weighting
spherical target
point cloud fitting
least squares
