A vision monitoring system for multipoint deflection of large-span bridge based on camera networking

This paper proposes a vision monitoring system for multipoint deflection of a large-span bridge that can well compensate the camera motion-induced errors. The camera network system (CNS) consists of a series of dual-camera stations that are linked with the cooperative markers. The infrared illumination supplement device is integrated in the camera station and the regular double-sided prism is selected as the cooperative marker to realize 24-h continuous monitoring. The feasibility and efficiency of CNS are verified by conducting a field test inside the steel box girders of a suspension bridge with a length of 1038 m. The results show that the proposed camera networking method can well solve the problem of the large-scale high-precision monitoring under unstable measuring platform. For the kilometer-level flexible bridge, CNS can meet submillimeter precision in quasi-static measurement, and can also provide deflection data similar to the conventional robotic total station and connecting pipe system in long-term monitoring. The geometric alignment of large-span bridges with dense multimeasuring points can be automatically monitored and recorded by CNS. The observed deflections of the studied suspension bridge are dominated with the vehicle load and wind load in the short term and are correlated with the temperature change in the long term. The proposed CNS can provide a new option for the bridge deflection monitoring and provide new parameters for bridge safety evaluation.