Time-domain model identification of structural dynamics from spatially dense 3D vision-based measurements

Single-Pixel Optical-Flow-Based Experimental Modal Analysis

“…5) is an ideally registered, sharp edge for two illumination cases. The desired effects are demonstrated using the response of SDOF system as an example, and due to the linearity of the Morlet wave integral (7), the same effects are expected for the MDOF systems. The first case represents the ideally illuminated edge where the light intensities through three pixels range linearly from 0 to 1.…”

“…The response of the structures contains several modes, and due to the linear property of the Morlet wave (7), each mode can be analysed separately [43,44].…”

“…The main advantage of using high-speed cameras in structural dynamics analyses is the non-contact and relatively simple measurement of the full-field structural response [1], typically providing several thousand measurement points. The advantage of such an over-determination of the response has been noted in several areas of structural dynamics, such as structural health monitoring [2], damage localisation [3], model updating [4], measurement [5] and reconstruction [6] of the sound-radiation field, and system identification [7]. In most of these areas, the analysis is made in the first step with displacement responses obtained by processing high-speed video images using the digital image correlation (DIC) method [8].…”

Computer Vision-Based Bridge Inspection and Monitoring: A Review

“…In addition, full-field mode shape identification has also been investigated. Willems et al [ 199 ] exploited the high spatial density of CV-based measurements to identify full-field mode shapes. Javh et al [ 200 ] used the Least-Squares Complex-Frequency method combined with the Least-squares Frequency-domain method to identify the high-frequency full-field mode shape of steel beams.…”