Research on steel structure point cloud skeleton extraction and construction accuracy evaluation based on BIM model assistance

Abstract In modern construction projects, the rapid development of 3D laser scanning technology and Building Information Modeling (BIM) technology has led to increasingly close integration of the two in engineering applications. By combining these two data sources, new methods for quality control in steel structure construction are provided. This paper proposes a method for extracting the skeleton of steel structures from point clouds, assisted by BIM models, and uses the refined skeleton data to evaluate construction accuracy. First, the collected 3D laser point cloud of the steel structure is registered using the BIM model. An original skeleton extraction algorithm is then employed to preprocess the steel structure point cloud, eliminating noise and redundant points to ensure data accuracy. Next, the denoised steel structure point cloud is combined with the designed BIM model to assess the accuracy of the actual construction. Experimental results show that this method not only effectively reduces noise in the measured point cloud and improves data quality but also accurately evaluates construction accuracy, providing reliable technical support for engineering quality control.