Full-Field Mode Shape Identification Based on Subpixel Edge Detection and Tracking

Most research on computer vision (CV)-based vibration measurement is limited to the determination of discrete or coarse mode shapes structure. The continuous edge structure in images has rich optical features, and thus, by identifying tracking movement structure’s edge, it possible determine high-resolution full-field without a preset target. present study proposes CV-based method shape identification using subpixel detection techniques. Firstly, Canny operator applied each frame video extract pixel-level edges, improved Zernike orthogonal moment (ZOM) technique adopted relocate precise edges. Then, all detected points are tracked obtain dense displacement time history that subsequently used frequencies compute combining covariance driven stochastic subspace (SSI-COV) with hierarchical cluster analysis. Finally, proposed verified aluminum cantilever beam laboratory Humen Bridge field. results show able detect more edges identify structures need for installing artificial targets advance, which provides valuable information structural condition assessment, especially small-amplitude vibrations.