Evaluating the use of digital image correlation for strain measurement in historic tapestries using representative deformation fields

Abstract

An analysis technique to assess the viability of digital image correlation (DIC) in tracking the full‐field strains across the surface of hanging historic tapestries is presented. Measurement uncertainty related to the use of the inherent tapestry image in tracking displacements is investigated through use of “synthetic” deformation fields. The latter are generated by mapping the details of a given tapestry image into finite element analyses. The combination of self‐weight loading, material non‐linearity, and image specific heterogeneity (related to slit stitching, damage, and patch‐restorations) serve to generate a bespoke deformation field complex enough to assess the reliability of DIC measurements. Accuracy is evaluated by comparing measured results with the original known deformations. The technique demonstrates that the optimum imaging settings and the choice of subset size for DIC analysis are strongly influenced by the tapestry image and the goal of the measurement, they are found using a compromise between conflicting objectives: minimising measurement error while maximising resolution.

Investigating mechanical damage mechanisms of tapestries displayed at different angles using 2D DIC

“…Also because of this, the resulting weaving structure is highly heterogeneous. The ends of the open slits are prime locations for high stress intensity factors [10] and are likely sites for failure due to creep. In general, the actual threshold stress of the wool yarns depends on different factors, including the environmental conditions.…”

Investigating the accuracy of digital image correlation in monitoring strain fields across historical tapestries

“…Comparing each deformed image directly with the reference image in order to track displacement Repetition An image containing one image repeating over and over could contain a lot of information and be isotropic but would be poor for DIC The user's choice of DIC settings and parameter values as well as the acquired image quality can significantly affect the accuracy of DIC measurements. The systematic approach developed in Alsayendoor et al, [17] and the current work provides a method for tapestry conservators using DIC techniques for strain monitoring, to determine optimum DIC parameters for optimised strain measurements prior to actual testing. Also, an approach to quantify error associated with DIC full field measurement using local strain maps is introduced considering image quality and DIC settings.…”

“…In the absence of ideal speckle pattern in using DIC techniques for full-field strain monitoring across historical tapestries inevitably leads to greater uncertainty in the experimental DIC-derived strain measurements; hence, a robust method is required to determine the viability and reliability of using DIC when using the tapestry's image and texture to track surface strains. As a continuation of Alsayednoor et al, [17] analysis tools have been further developed to aid conservators and conservation scientists determine the viability, accuracy and reliability of using DIC with tapestries, using the inherent tapestry image to track displacements. The analysis tools have the ability to determine the accuracy, potential sources of error and reliability of DIC measurements for any given tapestry image, by comparing the measured DIC results with a known finite element analysis (FEA)-generated synthetic representative deformation field.…”

“…Some are decided by the user in the DIC settings (see Table 2) relating to DIC algorithm accuracy, some are determined during the actual monitoring process and depend on issues such as image acquisition process like camera settings and lighting conditions (see Table 3), others depend on the suitability of the target image (see Table 4) for DIC-based tracking of displacements. In Alsayendoor et al, [17] the influence of just the subset size was explored, though many other sensitivity studies are possible including not just the DIC algorithm settings but also the image quality and even the FEA parameters such as mesh density/spatial resolution used to generate the synthetic deformation. Indeed, the only parameter that remains fixed is the suitability of the target image, that is, the inherent tapestry image is a fixed constraint.…”

Understanding the Role of Friction and Adhesion in the Display of Tapestries on Slanted Supports

“…Two cameras in a 3D system register the out of plane deformation while calculating the strain whereas 2D DIC reports out of plane displacements as errors (Alsayednoor et al, 2017). Alsayednoor, a researcher with the project, also demonstrated that variables such as the woven design of the tapestry can have an effect on the quality of the data recorded, while consideration of DIC methodology, such as the appropriate subset size, is also necessary (Alsayednoor et al, 2019).…”