Lightweight Defect Detection Equipment for Road Tunnels

The number of road tunnels has dramatically increased with extensive development of the transportation infrastructure, while the ever–increasing operational tunnels bring great pressure on the maintenance and management. Damages to the operational tunnels pose a significant risk to traffic safety. In particular, the lining cracks can significantly reduce structural load–bearing capacity of the tunnel. It is extremely important to develop a tunnel lining crack detection system to ensure operational and maintenance safety of the tunnel. Currently, the detection of lining cracks mainly relies on manual inspection, which is inefficient and cannot guarantee accuracy. Therefore, there is an urgent need to develop automatic lining crack detection technique to satisfy the growing demands in defect detection. To address the current issues regarding the lining crack detection, we develop a lightweight detection equipment for road tunnels in this study. Accurate lining images and posture information can be obtained through multi–sensor fusion, space–time synchronization and posture matching techniques. In addition, the YOLOv5 network is able to quickly identify damages from a large amount of data. This study improves the efficiency of crack detection, as well as addresses the issues on the large volume and inconvenience of road closure during detection of the traditional detection equipment, which can serve as a useful reference for tunnel operation and maintenance.