SFDM: Structural full-field dynamic displacement monitoring utilizing computer vision-enhanced targetless monitoring system

Full-field dynamic displacement monitoring with high spatial resolution provides a more comprehensive representation of the dynamic characteristics of bridge structures, thereby significantly enhancing the effectiveness of bridge structural health monitoring (BSHM). In this study, a structural full-field dynamic displacement monitoring (SFDM) method is proposed to overcome the limitations of traditional discrete-point measurement techniques. To achieve targetless and accurate extraction of the lower edge of bridge structures, an effective computer vision framework is developed by integrating a robust fusion image processing (RFIP) strategy with a row-wise adaptive edge detection (RAED) approach. Additionally, a multi-target displacement transformation model (MDTM) is proposed to convert subpixel-level image displacements into physical displacements. A tied-arch bridge field test was conducted to evaluate the performance of the proposed method. The experimental results indicate that the proposed SFDM achieved correlation coefficients higher than 0.964, while the corresponding RMSE and NRMSE remained below 0.282 mm and 9.178% across all test cases, respectively, thereby confirming its accuracy and effectiveness for full-field displacement monitoring in BSHM applications.