一种顾及主方向的点云配准施工检测方法  

Construction Detection Method for Point Cloud Registration Considering the Main Direction

作　　者：唐菲菲 王巨晖 王章朋 陈星池 TANG Feifei;WANG Juhui;WANG Zhangpeng;CHEN Xingchi(School of Smart City,Chongqing Jiaotong University,Chongqing 400074,China;Shanghai Baoye Group Co.,Ltd.,Shanghai 201908,China)

机构地区：[1]重庆交通大学智慧城市学院,重庆400074 [2]上海宝冶集团有限公司,上海201908

出　　处：《地理空间信息》2023年第12期43-46,共4页Geospatial Information

基　　金：重庆交通大学研究生科研创新项目(2022S0087);重庆市教委科学技术研究项目(KJQN201900728)。

摘　　要：面向建筑物施工的全局偏差检测需求,提出一种顾及主方向的PCA-ICP点云配准施工检测方法。首先将BIM设计模型数据转换为BIM点云数据;然后运用顾及主方向的PCA(主成分分析)粗配准结合ICP(迭代最近点)精配准,即顾及主方向的PCA-ICP配准方法,在无控制点条件下实现三维激光点云与BIM点云配准,并计算配准均方误差E;最后利用M_(3)C_(2)算法和均方误差E计算出三维激光点云与BIM点云之间的偏差值和偏差值误差范围,最终统计得到施工偏差结果。以某体育场馆钢结构为例,实验结果表明,该方法较常规偏差检测方法可以得到建筑物实际施工的全局偏差。实验结果可为建筑物施工提供参考。Aiming at the global deviation detection requirements of building construction,we proposed a PCA-ICP point cloud registration con-struction detection methodconsidering the main direction.Firstly,we converted the BIM design model data into BIM point cloud data.Then,we used PCAcoarse registration with respect to the principal direction combined with ICPfine registration,that is,taking into account the PCA-ICP registration method in the main direction to realize the registration of 3D laser point cloud and the cloned laser point cloud under the condition of no control points,and calculated the registration mean square error.Finally,we utilized the M_(3)C_(2) algorithm to calculate the deviation value under the premise of considering the mean square error and obtained the construction deviation result by statistics.Taking the steel structure of a stadi-um as an example,we found that the method can obtain the global deviation of actual construction of building compared with the conventional deviation detection method,providing a reference for building construction.

关 键 词：光点云 点云配准 点云主方向 M_3C_2距离计算 施工检测 

分 类 号：P258[天文地球—测绘科学与技术]