Vision-based systems for structural deformation measurement: case studies

Abstract

Vision-based systems offer a promising way for displacement measurement and receive increased attention in civil structural monitoring. However, the working performance of vision-based systems, especially the measurement accuracy and the robustness to different field conditions is not fully understood. This study reports three cases studies of vision-based monitoring tests including one in a laboratory, one on a short-span bridge and one on a long-span bridge. The tracking accuracy is quantified in laboratory conditions in the range of 0.02 pixel to 0.20 pixel depending on the target patterns as well as the tracking method selected. The measurement performance under several field challenges are investigated including long-range measurement (e.g. camera-to-target distance at 710 m), low-contrast target patterns, changes of target patterns and changes in lighting conditions. Three representative tracking methods for the video processing, i.e. correlation-based template matching, Lucas Kanade (LK) optical flow estimation and scale-invariant feature transform (SIFT) were used for analysis, indicating their advantages and shortcomings for field measurement. One of the main observations in field application is that changes in lighting conditions might cause some low-frequency measurement error that could be misunderstood without the prior knowledge about structural loading conditions.