Basis-updating for data compression of displacement maps from dynamic DIC measurements

Abstract

The extraction of useful information and removal of redundant noise from data has become a major research topic in recent years. Data compression is necessary for all kinds of analysis, and the demand for efficient compression techniques has gained much attention. Digital image correlation is a camera-based measuring system, which has been widely applied in strain analysis because of the convenience of measuring displacement fields by simply selecting a region of interest. Currently, there is interest in applying such methods to engineering structures in dynamics. However, one of the major issues related to the integration of camera-based systems with dynamic measurement is the generation of huge amounts of data, typically extending to many thousands of data points, because of the requirements of high sampling rate, spatial resolution, and long duration of recording. In this paper a new algorithm is presented that addresses the need for efficiency in full-field data processing. By making use of the data itself and combining the concept of sparse representation with Gram-Schmidt orthogonalisation, the number of basis function used to represent the data can be reduced and a concise decomposition established. In both simulated and experimental cases, the compression ratios for data size and number of signals used in operational modal analysis are substantially diminished, thereby demonstrating the effectiveness of the proposed algorithm. A reduced number of new basis functions is determined for the representation of data under the condition that the reconstructed displacement map reproduces the raw measured data to within a chosen threshold on the coefficient of correlation.

Compressed sensing for OMA using full-field vibration images

“…Classical polynomials, typically Zernike and Tchebichef, and AGMD kernels are all preselected independently of the image in question and therefore do not, in general, provide the most economical reduction. The present authors developed a basis updating procedure [37] inspired by the K-SVD algorithm [38].…”

“…Following the CS recovery, operational modal analysis of the SD data was carried out according to the procedures described in §2 (and in [37]). Modal identification requires only the signals in the SD domain, then mode-shape visualisation can be achieved using the kernel functions.…”

“…The advantages of this approach were exemplified in the vibration study of a Fiat Panda bonnet liner, having a complicated curved surface and holes, with fullfield modal testing carried out in the SD (AGMD) domain -rather than in the physical domain of a limited number of acceleration sensors. Recent research by Chang et al [37] addresses the fact that the pre-defined AGMD, though general, is not necessarily the most efficient (or compact) basis. They presented an algorithm inspired by K-SVD which uses measurements to directly update the chosen kernel functions to obtain a more representative and concise reconstruction of full-field data.…”

Vibration Measurement of a Metal Sheet Using Single-Camera Digital Image Correlation with Projection Components

“…Huňady and Hagara [14] realized a huge efficiency improvement using the enhanced frequency function (EFRF) instead of FRF in modal parameter estimation procedure, with hundreds or thousands of output degrees of freedom in full-field modal analysis being necessary. Chang et al [32] followed the adaptive geometric moment descriptor (AGMD) and combined it with K-SVD and Gram-Schmidt orthonormalization (GSO) to achieve data compression of displacement maps. Passieux et al [33] developed a new regularized DIC method for time-dependent measurements to improve the space field uncertainties and achieve a trade-off between the frame rate and spatial resolution.…”

Full-Field Operational Modal Analysis of an Aircraft Composite Panel from the Dynamic Response in Multi-Impact Test

“…In [ 37 ], the authors performed the first integration of DIC and OMA for the analysis of a helicopter rotor blade during operation using the Ibrahim time-domain method. Considering the amount of data generated by full-field measurement, the evaluation of a compressed DIC sensing methodology has been performed for OMA using the stochastic subspace identification time-domain method [ 38 , 39 ]. In [ 40 ], the authors performed an OMA of the flexible blades of a two-bladed rotor using DIC measurements.…”