Accurate Deformation Monitoring on Bridge Structures Using a Cost-Effective Sensing System Combined with a Camera and Accelerometers: Case Study

Abstract

Information on deformation is critical for bridge condition evaluation, but accurate characterization, usually via discrete displacement measurements, remains a challenging task. Vision-based systems are promising tools, possessing advantages of easy installation, low cost, and adequate resolution in time and frequency domains. However, vision-based monitoring faces several field challenges and might fail to achieve the required level of working performance in some real-world test conditions (e.g., involving low-contrast patterns and mounting instability of optical sensors). To make the best use of the potential of vision-based systems, a mixed sensing system consisting of a consumer-grade camera and an accelerometer is proposed in this study for accurate displacement measurement. The system considers automatic compensation of camera shake and involves an autonomous data fusion process for noise reduction. The proposed system is demonstrated by a field monitoring test on a short-span railway bridge and is validated to offer higher accuracy and wider frequency range than using a camera alone. Displacement data by the mixed system are demonstrated to be viable for estimating bridge influence line (IL), indicating the potential for bridge condition assessment.