基于地面三维激光扫描与ICP算法的建筑点云配准方法  

Building point cloud registration method based on ground 3D laser scanning and ICP algorithm

作　　者：黄俊龙 HUANG Junlong(Xinzhou Urban and Rural Planning and Design Institute Co.,Ltd.,Xinzhou 034000,China)

机构地区：[1]忻州市城乡规划设计院有限责任公司,山西忻州034000

出　　处：《经纬天地》2025年第1期18-23,共6页Survey World

摘　　要：针对迭代最近算法在大型建筑物点云精配准时存在的计算效率以及配准精度低等缺陷,分析了基于地面三维激光扫描技术的建筑点云模型,对传统迭代最近算法中的候选点选取、同名点赋权以及误差函数选取等进行了改进。结果表明:改进后的迭代最近算法的均方根误差较传统迭代最近算法提高了59.12%,其最短配准耗时也远低于传统算法,相较于传统迭代最近算法,其效率提高了61.88%,极大提高了算法配准精度以及配准效率。在建筑工程测量领域有重要的应用价值。Aiming at the existed shortcomings as low computational efficiency and registration accuracy of the iterative nearest algorithm in the fine registration of point clouds of large buildings,we analyse the point cloud model of buildings based on terrestrial 3D laser scanning technology,and improve the selection of candidate points,assignment of homonymous points and the selection of error function in the traditional iterative nearest algorithm.The results show that the root mean square error of the improved iterative nearest algorithm is 59.12%higher than that of the traditional iterative nearest algorithm,and its shortest registration time is much lower than that of the traditional algorithm.Compared with the traditional iterative nearest algorithm,its efficiency is 61.88%higher,which greatly improves the algorithm's registration accuracy and registration efficiency.This method has important application value in the field of construction engineering measurement.

关 键 词：建筑工程 激光扫描仪 ICP 点云 精配准 

分 类 号：P237[天文地球—摄影测量与遥感]