Review of Computer Vision-based Structural Displacement Monitoring

Research and application advances of computer vision-based structural displacement monitoring methods were addressed in this paper. Furthermore, overviews of four aspects, system configuration, computing methods, influence factors, and real applications, were provided. The current research limitations of computer vision-based structural displacement monitoring were explored and reasonable suggestions were provided. With respect to system configuration, the selection criteria and merits and demerits of various cameras, lenses, and markers were presented. Suggestions regarding selection of cameras and lenses were provided, and different scenarios of manual markers were presented. With regard to computing methods, the realization methods of camera calibration, feature extraction, target tracking, and displacement calculation were introduced. Additionally, the measurement error induced by lens distortion and the necessity of camera calibration were analyzed, and two practical and simplified camera calibration methods were introduced. Regarding influence factors, the sources of systematic measurement errors caused by hardware, image processing algorithms, and environmental factors were analyzed, and solutions for mitigation of systematic errors were elaborated. With regard to real applications, the applications of computer vision-based structural displacement monitoring in structural condition assessment were introduced, which included structural behavior analysis, loading capacity evaluation, modal parameter identification, model updating, damage detection, and cable force estimation. © 2019, Editorial Department of China Journal of Highway and Transport. All right reserved.