Algorithm development for automatic detection of progressive damage in tunnel cross-sectional geometry

Saved in:

Bibliographic Details
Published in:	ISARC. Proceedings of the International Symposium on Automation and Robotics in Construction vol. 42 (2025), p. 1494-1501
Main Author:	Nuñez, Christopher
Other Authors:	Regalado, Marck
Published:	IAARC Publications
Subjects:	Outliers (statistics) Damage detection Algorithms Tunnels Deformation analysis Data acquisition Weak rock Monitoring Image segmentation Structural integrity Outliers (landforms) Datasets Photogrammetry Scanners Lasers Civil engineering Data processing Unmanned aerial vehicles Automation Deformation Geometry Robotics Geology
Online Access:	Citation/Abstract Full Text - PDF
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Abstract:	Accurate assessment of tunnel conditions is vital for ensuring long-term safety and structural integrity, particularly in weak rock masses with complex geological conditions. This study introduces a methodology to analyze tunnel deformation and potential damage using point cloud data acquired in 2019, 2022, and 2024 through terrestrial laser scanning (TLS). The datasets were processed to remove outliers, irrelevant elements, and internal features, resulting in a clean representation of the tunnel's geometry. A novel algorithm integrates point cloud segmentation, centerline alignment, and projection techniques to derive tunnel profiles and monitor structural changes over time. The results demonstrate TLS's effectiveness in capturing high-resolution data for ovalization monitoring, revealing a consistent reduction in cross-sectional area throughout the tunnel and a rightward shift in the 2024 profiles. This approach underscores the potential of TLS for precise structural monitoring and deformation analysis in challenging environments.
Source:	Advanced Technologies & Aerospace Database