The work performance of stereo digital image correlation (stereo-DIC) technologies, especially the operating accuracy and reliability in field applications, is not fully understood. In this study, the key technologies of the field remote 3D displacement sensing of civil structures based on stereo-DIC have been proposed. An image correlation algorithm is incorporated in improving the matching accuracy of control points. An adaptive stereo-DIC extrinsic parameter calibration method is developed by fusing epipolar-geometry-based and homography-based methods. Furthermore, a reliable reference frame that does not require artificial markers is established based on Euclidean transformation, which facilitates in-plane and out-of-plane displacement monitoring for civil structures. Moreover, a camera motion correction is introduced by considering background points according to the camera motion model. With an experiment, the feasibility and accuracy of the proposed system are validated. Moreover, the system is applied to sense the dynamic operating displacement of a 2 MW wind turbine’s blades. The results show the potential capability of the proposed stereo-DIC system in remote capturing the full-field 3D dynamic responses and health status of large-scale structures.
Read full abstract