Phase-based Displacement Measurement using an Optimal Designed Filter and Motion Amplified Video-based Structural Damage Detection

Compared to conventional contact techniques, advanced displacement measurement techniques using computer vision have shown several advantages, such as high spatial resolution and no mass loading effect. Recently developed phase-based motion estimation techniques can be used to accurately measure motion signals from the local phase without speckle patterns on a structure. However, various conditions, such as complex illumination and insufficient lighting, can generate significant errors during the measurement process. In this study, an optimally designed steerable filter was used to provide a stable local phase, resulting in significant correlation with the motion in the phase-based optical flow technique. For structural health monitoring and damage detection, the amplified video was automatically generated using image processing and the proposed measurement technique. Several experiments were performed on a five-story structure to validate the proposed techniques using accelerometers, a laser Doppler vibrometer, and a camera. The experimental results demonstrated that the proposed techniques could be used to accurately measure displacement and identify the modal frequencies and full-field operational deflection shape. In addition, bolt-loosened damage was successfully detected using the mode-amplified video and operational curvature shape.