Vibration-based damage imaging via high-speed cameras with 3D digital image correlation using wavelet transform

Vibration-based damage detection (VBDD) techniques are based on sensing and then analyzing the structure dynamic response to detect damage(s) in structures. Numerous techniques have been used to sense the dynamic response of the structures followed by VBDD techniques to locate and size the damage(s). High speed cameras combined with digital image correlation (DIC) software provides a viable solution since cameras act as image sensor acquiring high fidelity full-field images and multi-axial stereo measurement remotely without added mass to the structure. In this paper, an aluminum plate with a rectangular damage area was excited by a mechanical shaker and operating deflection shape (ODS) was taken to examine the higher-order vibration modal response for a frequency band ranging from 1 Hz to 1 kHz. Both in-plane and out-of-plane displacements are acquired for further analysis. With this enriched information, displacement can be interpreted in complex domain to discuss the domination over each modes and rotational response behavior. In order to isolate minute displacement perturbation induced by damages from the full-field measurement, mode shape curvature, simply called modal curvature (MC), was taken with two-dimensional wavelet transform (2-D CWT) acting like a Laplacian filter as the mother wavelet in time-frequency domain for damage feature detection. The combination of this proposed WT-MC method to be noise tolerant and characterize mode shape irregularities induced by hidden damages. The first 12 vibration modes were captured and processed by WT-MC with weighting summed average and the damage imaging closely resemble the actual damage area.