ABSTRACT
The fatigue cracks occurs in the orthotropic steel deck since the cumulative damage caused by the heavy vehicles, which reduce the serviceability of the bridge. Thus, the inspection and reinforcement of fatigue cracks is significant in the bridge maintenance. In this paper, we proposed a machine-vision based system for the monitoring of fatigue cracks in U rib-to-deck weld seams, including both hardware and software. To be specific, the image acquisition device is firstly developed containing three major modules. Then an initial image calibration method is innovated for obtaining a measurable panoramic fatigue crack image. Furthermore, a deep learning-based crack recognition algorithm is designed to segment the crack areas. Finally, the features of a crack are obtained by applying the image processing techniques, involving its area, length, width and direction. Finally, a field experiment was carried out on a real steel suspension bridge. The corresponding results validate its high reliability and efficient in monitoring of a fatigue crack around U-ribs. This work is not only of practical value to the management and maintenance of the OSG bridges in engineering, but also critical for the mechanism researches on fatigue cracks propagation.
