基于DLG的LiDAR点云自动配准及精度分析  

Automatic Registration and Precision Analysis of Laser Point Cloud Data Based on DLG

作　　者：胡思雨 代侦勇[1] 

机构地区：[1]武汉大学资源与环境科学学院,湖北武汉430070

出　　处：《地理空间信息》2018年第2期34-36,39,共4页Geospatial Information

摘　　要：针对LiDAR数据平面精度达不到大比例尺传统测绘要求的问题,提出一种基于DLG的LiDAR点云自动配准方法。通过提取点云地物边界曲线与DLG数据曲线进行匹配,得到校正向量,实现机载LiDAR点云的全地形自动配准。本方法引入傅里叶描述子用于描述曲线特征,克服了常规方法中只提取直线特征的局限。通过对配准结果的精度统计和分析表明,能够达到较高的配准精度。This article presented a method of automatic registration of airborne Li DAR data. Firstly, the paper extracted the boundary from point clouds by filtering algorithms with slope factors. And then, the paper described the characteristics of the curve by Fourier descriptors, and matched the boundary of DLG to obtain correction vector. Finally, the paper got the correction vector of any points by spatial interpolation, and achieved the automatic registration of point clouds of all kinds of terrain. The main advantage of this method is the description of curve feature by Fourier descriptors, which overcoming the defect of extracting the straight line of buildings only. The experimental results show that the new method can obtain more accurate results and meet the requirements of traditional geomatics.

关 键 词：LIDAR 激光点云 DLG 傅里叶描述子 自动配准 

分 类 号：P237[天文地球—摄影测量与遥感]