Multi-Point Displacement Synchronous Monitoring Method for Bridges Based on Computer Vision

Abstract

Bridge displacement is an important part of safety evaluations. Currently, bridge displacement monitoring uses only a few measurement points, making it difficult to evaluate safety. To address this problem, we propose a multi-point displacement synchronous monitoring method. The structural surface has abundant natural texture features, so we use the feature points of the structural surface as the displacement measurement points and propose a feature point displacement calculation method. Furthermore, we conduct experiments on a beam in the laboratory and obtain the beam’s multi-point displacement monitoring results. The monitoring results show that the displacement of some feature points is mismatched. We propose the use of the structural deflection curve to eliminate the feature point displacement mismatches. This method uses the maximum rotation angle of the deflection curve to eliminate displacement mismatches. The results indicate that it is effective to eliminate displacement mismatches in simple structures, such as simply supported beams. Finally, we obtain the test beam’s multi-point displacement synchronous monitoring results. Compared with the 3D laser scanning measurement method, the maximum error of the monitoring results is 8.70%. Research shows that the main reason for the monitoring error is image noise, and the noise interference problem due to its application in practical bridges requires further investigation. Compared with traditional displacement monitoring, this method has significant economic, efficiency, and data integrity advantages. The method has application prospects for multi-point displacement monitoring of simple structures, such as simply supported beams.

Keywords

• structural health monitoring
• displacement monitoring
• deformation monitoring
• feature point detection
• feature extraction
• deflection curve