Project Title

UAV-enabled Structure-From-Motion Photogrammetry for Bridge Crack Detection and Characterization

Researcher(s)

Xiong Zhang, PI, Missouri University of Science & Technology
Jenny Liu, Co-PI, Missouri University of Science & Technology
Genda Chen, Go-PI, Missouri University of Science & Technology

Project Description

Bridge is a common structure form widely adopted in the engineering construction, which plays an important role in traffic and transportation system. Crack is considered as an indicator for a bridge’s structural and functional failures, and crack detection is one of the major tasks during bridge inspection to maintain the structure health and serviceability of a bridge. Literature review indicated that until now detection of 3D bridge crack is still a great challenge for structural engineer and there is little research on the automatic characterization of 3D bridge cracks. The objective of this proposed research is to develop a UAV-Enabled Structure-From-Motion Photogrammetry for detection and characterization for 3D bridge crack detection. Commercially available low-cost UAVs will be used to take the images needed for the analyses. A Structure-From-Motion Photogrammetry algorithm will be developed to reconstruct the 3D models of the bridges and deep learning will be used to automatically determine the cracks from the 3D modes. With the 3D models, crack characterization such as crack lengths, depths, widths, and patterns on bridge components can be automatically measured with high accuracy. The crack measurements will be compared and validated against results obtained from other existing methods such as local sensors and the high accuracy LiDAR system. This method for 3D crack mapping will provide us a high accuracy, low cost, and easy-to-operate tool for bridge maintenance and management.

Project Details

Project Visuals