基于PCA方法的BIM构件重建与误差分析  

BIM Component Reconstruction and Error Analysis based on PCA Method

作　　者：漆征鹏 周拥军[1,2] QI Zhengpeng;ZHOU Yongjun(Schoo1 of Naval Architecture,Ocean and Civil Engineering,Shanghai Jiaotong University,Shanghai 200240,China;Shanghai Key Laboratory for Digital Maintenance of Buildings and Infrastructure,Shanghai 200240,China)

机构地区：[1]上海交通大学船舶海洋与建筑工程学院,上海200240 [2]上海市公共建筑和基础设施数字化运维重点实验室,上海200240

出　　处：《粉煤灰综合利用》2022年第2期1-9,共9页Fly Ash Comprehensive Utilization

基　　金：上海社发基金(20dz1201300)。

摘　　要：本文针对具有规则形状的建筑构件的点云提取和几何重建问题,提出了主成分分析(PCA)与奇异值分解(SVD)相结合的构件分析方法。先对完成粗提取的构件点云计算散度矩阵,再利用PCA方法得到的特征值和特征向量来分析构件形状和方位,最后考虑点云的完整性、测量精度及采集密度等因素对该方法的稳定性和精度进行了仿真分析。结果表明:本文方法能有效分析构件的位置、形状和方位,点云的密度、精度和完整性对分析结果的影响较小,可广泛应用于基于点云的建筑构件BIM几何建模。Aiming at the problem of point cloud extraction and geometric reconstruction of building components with regular shapes,the paper proposes a component analysis method combining Principal Component Analysis(PCA)and singular value decomposition(SVD).Firstly,the divergence matrix is calculated for the component point cloud that has been roughly extracted,and then the shape and orientation of the component are analyzed by using the eigenvalues and eigenvectors obtained by the PCA method.The stability and accuracy of the model were simulated and analyzed.The results show that the method in this paper can effectively analyze the position,shape and orientation of the components,and the density,accuracy and integrity of the point cloud have little influence on the analysis results,and can be widely used in the BIM geometric modeling of building components based on the point cloud.

关 键 词：点云 BIM 主成分分析 奇异值分解 

分 类 号：TU3[建筑科学—结构工程]