Abstract:Structural Health Monitoring, Ahead of Print. This paper explores a full-field non-contact optical sensing technique using a stereo camera for imaging hidden damage based on vibration-based damage detection methodology in structural health monitoring. The technique utilizes a pair of digital cameras to capture dynamic operational deflection shapes (ODSs) over the region of interest (ROI) of a structure's surface via digital image correlation (DIC) when subjected to vibrational excitation. This research overcomes bottlenecks in using high vibration modes for imaging the hidden damage area by (1) applying DIC to operational modal analysis with simple pick-peaking techniques to gather natural frequencies and operational mode shapes in plate structures, while (2) using wavelet analysis to reveal the image of the damage region as a means for baseline-free global damage quantification. In the feasibility study, four cases with two aluminum plates with large damage regions were investigated with a vibration shaker generating a frequency sweep up to 1 kHz. The stereo camera imaged the speckled surface of the plate with white light. Once the dynamic ODSs in the ROI were observed using DIC, the natural frequencies and associated operational mode shapes were extracted using a peak-picking technique in the frequency spectrum. Natural frequencies and operational mode shapes from finite element analysis correlated well with the experimental observations from three-dimensional DIC for all 12 vibration modes respectively. A wavelet transform mode shape curvature (WT-MSC) technique to obtain the modal shape curvature via a two-dimensional continuous wavelet transform with a Mexican Hat analyzing wavelet was then implemented on each of the first 12 vibration mode shapes. A damage image condition that incorporates all weighted wavelet coefficients is proposed to image the damage region. The hidden damage was visualized clearly with WT-MSC, as the technique is much less sensitive to noise than the use of MSC alone, and the use of high vibration modes exhibiting larger mode shape curvatures provided a greater sensitivity for imaging the damage region. Hidden damage regions were successfully visualized in all four cases.

Prior-shape-guided Photometric Stereo Model for 3D Damage Measurement of Worn Surfaceiks

Model-informed Deep Learning Strategy with Vision Measurement for Damage Identification of Truss Structures

In-situ 3D reconstruction of worn surface topography via optimized photometric stereo

Distortion Correction in Surface Defects Evaluating System of Large Fine Optics

Prior surface model guided dense reconstruction via graph-cuts

Wear Depth Estimation from Single 2-D Image Based on Shape from Shading and Convolutional Neural Network Hybrid Model for In-Situ Wear Assessment

Image-Based Hidden Damage Detection Method: Combining Stereo Digital Image Correlation and Finite Element Model Updating

Quantitative evaluation for small surface damage based on iterative difference and triangulation of 3D point cloud

A Numerical Simulation Procedure for Evaluating the Accuracy of 3-Dimensional Photogrammetric Models and Its Application to Geometric Parameters of Discontinuities in Rock Masses

Worn area modeling for automating the repair of turbine blades

Lambertian Reflection Separation under High Reflectiveness for Worn Surface Reconstruction with Insufficient Samples.

Vibration-based hidden damage imaging using stereo cameras with digital image correlation

Optical Subsurface Damage Evaluation Using LSCT

Three-dimensional Reconstruction of Wear Particle Surface Based on Photometric Stereo

Concrete Surface Damage Volume Measurement Based on Three-Dimensional Reconstruction by Smartphones

Surface damage quantification of postearthquake building based on terrestrial laser scan data

A Robust 3-D Abrasion Diagnosis Method of Pantograph Slipper Based on Stereo Vision

Confocal laser scanning and 3D reconstruction methods for the subsurface damage of polished optics

Comparison-embedded Evidence-Cnn Model for Fuzzy Assessment of Wear Severity Using Multi-Dimensional Surface Images

3-D Particle Surface Reconstruction from Multiview 2-D Images with Structure from Motion and Shape from Shading (january 2020)

Holistic image‐based analysis of damage on concrete surfaces—A multifaceted approach based on supervised machine learning