摘要
截至2023年,我国城镇化率已达66.16%,老旧小区改造更新迫在眉睫。相较于拆除重建,有机更新更具优势,于是BIM逆向建模方法成为本研究主要探讨的问题。提供一种针对点云数据处理算法的关键步骤,包括点云标靶配准、点云格式及转换、K维树(K-Dimensional树,简称kd-tree)、法线估计和点云下采样、统计滤波和球旋转算法(BPA)表面重建。在点云标靶配准中,给出了标靶布设距离、重叠度要求以及配准参数的求解过程。kd-tree是用于在k维空间组织点的数据结构,用以完成混合搜索。法线估计通过协方差分析计算每个点的法线。点云下采样利用体素下采样方法以减少数据量、去除噪声点,提高建模精度和效率。通过这种算法,可以更好地进行表面重建和模型拟合,提高模型构建精度和质量。
As of 2023,China′s urbanization rate has reached 66.16%,indicating a pressing need for the renovation and renewal of existing residential areas.In comparison to demolition and reconstruction,organic renewal presents a more advantageous approach.Consequently,the BIM reverse modeling method has emerged as a central topic of discussion in this article.This article primarily presents a pivotal stage in the processing of point cloud data,encompassing point cloud target registration,point cloud format and conversion,K-dimensional tree(kd-tree),normal estimation and point cloud downsampling,statistical filtering,and ball rotation algorithm(BPA)surface reconstruction.In point cloud target registration,the process of determining the target deployment distance,overlap requirements,and registration parameters is provided.A kd-tree is a data structure that is used to organize points in a k-dimensional space for the purpose of performing a mixed search.Normal estimation involves calculating the normal of each point through the use of covariance analysis.Point cloud downsampling makes use of voxel downsampling methods with the objective of reducing the data volume,removing noise points,and improving the accuracy and efficiency of modeling.Through the implementation of this algorithm,surface reconstruction and model fitting can be performed in a more effective manner,thereby enhancing the precision and quality of the model construction.
作者
汪成
WANG Cheng(China Railway Construction Group Co.Ltd.,Beijing 100040,China)
出处
《铁道建筑技术》
2025年第4期188-192,共5页
Railway Construction Technology
基金
中铁建设集团有限公司科技研发计划项目(2023-32c)。
关键词
老旧小区改造更新
BIM逆向建模
点云数据处理算法
点云标靶配准
点云格式转换
法线估计
renovation and renewal of old residential areas
BIM reverse modeling
point cloud data processing algorithm
point cloud target registration
point cloud format conversion
normal estimation
