Deep learning-based pipe segmentation and geometric reconstruction from poorly scanned point clouds using BIM-driven data alignment

• Published in: Automation in construction Vol. 173; p. 106071
• Main Authors: Yu, Wanchen, Shu, Jiangpeng, Yang, Zihan, Ding, Hongliang, Zeng, Wuhua, Bai, Yong
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.05.2025
• Subjects: Alignment strategies; BIM-generated datasets; Geometric reconstruction; Pipe; Point cloud segmentation; Point cloud segmentation; Alignment strategies; Geometric reconstruction; Pipe; BIM-generated datasets
• ISSN: 0926-5805
• DOI: 10.1016/j.autcon.2025.106071

Pipe reconstruction is an important prerequisite for pipe maintenance. However, scanned point clouds often contain defects, presenting a significant challenge for automated segmentation and geometric reconstruction. To address this challenge, this paper proposes a learning-based segmentation method, PipeSegNet, along with a geometric reconstruction process. In the segmentation stage, a method is developed to generate datasets with controlled density from BIM. Meanwhile, alignment strategies are introduced to address feature and label inconsistencies between BIM-generated and real datasets. PipeSegNet enhances global and local perceptual capability, achieving pipe segmentation accuracy of 96.37 % and IoU of 91.45 %, ensuring high-quality reconstruction. Comparative and module evaluation experiments demonstrate the effectiveness of PipeSegNet combined with the alignment strategies. The total average relative error of the reconstructed pipes is 2.73 %. This paper provides valuable insights into the pipe segmentation and reconstruction from point clouds, particularly in scenes with poor scanning quality, contributing to efficient infrastructure maintenance. •Propose PipeSegNet, a point cloud segmentation method that enhances both global and local feature perception.•Develop a precise pipe geometric reconstruction algorithm for poorly scanned point clouds.•Introduce a method for generating high-quality, well-structured point cloud datasets from BIM with controlled density.•Implement feature and label alignment strategies to leverage BIM-generated data for improving PipeSegNet's performance.