Image-based crack assessment of bridge piers using unmanned aerial vehicles and three-dimensional scene reconstruction

Tackles two problems in UAV crack inspection of bridge piers: close-up photos combine perspective and curved-surface distortion, and each shot covers so little of the pier that crack location is unclear. It combines flight path and photography strategy, crack detection and 3D reconstruction into one workflow. Projecting detected cracks onto a triangulated 3D mesh corrects both distortions and places each crack on the pier, and the method was validated with error analysis on a real pier.

Key findings

• Crack projection onto a 3D triangular mesh corrects perspective and curved-surface distortion together
• Workflow outputs crack width and a located 3D surface model of the pier
• Validated with field tests and error analysis on a real bridge pier

Limitations

• Abstract only reviewed; numeric error values not given in the abstract
• Single pier field test