Mesh-based digital image correlation method using higher order isoparametric elements

Abstract

By using the 8-node-quadrangular isoparametric element to discretize the speckle image, a 8-node-quadrangular mesh-based digital image correlation method is proposed. The model of 8-node-quadrangular mesh digital image correlation and its solving scheme based on the Newton–Raphson iteration have been constructed. The method is realized and then verified with speckle images produced by numerical simulations and real experiments. The results show that the 8-node-quadrangular mesh digital image correlation has a similar accuracy on a simple deformation measurement (translation, rotation and constant strain deformation) compared with mesh digital image correlation using the simple 4-node-quadrangular element. But for a heterogeneous deformation field measurement, the 8-node-quadrangular mesh digital image correlation has a higher accuracy because the 8-node isoparametric element used has the second-order interpolation accuracy, which is higher than that in 4-node-quadrangular mesh digital image correlation and in the conventional subset-based digital image correlation. Moreover, the 8-node-quadrangular mesh digital image correlation has the advantage for the measurement of a specimen with circular edges; the 8-node-quadrangular mesh digital image correlation can measure the displacement of all points close, even on, the curve edges because the 8-node-quadrangular isoparametric element is able to be mapped onto an element with second-order curved edges.