Tunnel Deformation Inspection via Global Spatial Axis Extraction from 3D Raw Point Cloud.

Area of Science:

• Geotechnical Engineering
• Civil Engineering
• Geomatics Engineering

Background:

• Global inspection of large-scale tunnels is crucial for structural stability and driving safety.
• LiDAR scanners provide 3D point clouds for tunnel deformation inspection (TDI).
• Raw LiDAR data often suffers from occlusions, missing data, and noise, hindering accurate analysis.

Purpose of the Study:

• To develop an effective algorithm for tunnel deformation inspection using poor-quality 3D point clouds.
• To accurately extract the global spatial axis of a tunnel from noisy LiDAR data.
• To enable automated computation of tunnel deformation parameters.

Main Methods:

• Proposed an iterative fitting optimization algorithm to extract the tunnel's spatial axis.
• Generated normal planes in the Frenet frame to create cross-sections from the point cloud.
• Fitted circles to cross-sections and approximated centers with a B-Spline curve iteratively until convergence.
• Designed a segmentation approach for cross-sections to compute deformation parameters.

Main Results:

• Successfully extracted the accurate spatial axis of tunnels even from poor-quality point clouds.
• Demonstrated the feasibility and effectiveness of the proposed tunnel deformation inspection method through experiments.
• Enabled automated computation of various tunnel deformation inspection parameters.

Conclusions:

• The developed algorithm effectively extracts the spatial axis from noisy tunnel point clouds.
• The method provides a robust solution for tunnel deformation inspection, enhancing structural safety.
• This approach facilitates automated and accurate assessment of tunnel integrity.