Calibration methodology of a vision system for measuring the displacements of long-deck suspension bridges

摘要: SUMMARY Structural health monitoring is an emergent powerful diagnostic tool that can be used to identify and prevent possible failures of the various components comprise infrastructure. In a suspension bridge, measurement vertical transversal displacements plays important role for its safety evaluation. Considering restrictions usually found on these structures, enhanced solution comprises non-contact vision-based system with dynamic response, accuracy amplitude range well suited physical phenomenon measure. A methodology perform vision calibration described, which carried out in situ, while deck moving, requiring little effort minimum set information. The key idea use factorization method get initial estimation object shape camera's parameters, incorporate knowledge about distances between targets non-linear optimization process achieve metric reconstruction optimize parameters estimation. Results obtained by numerical simulation experiments are reported, showing negligible influence noisy data performance also robustness estimation. Tests assess tracking bridge motion showed that, even environmental severely affected noise, it measure excellent levels accuracy. Copyright © 2011 John Wiley & Sons, Ltd.