Digital image correlation (DIC): Decorrelation problem due to large strains

Abstract

The digital image correlation (DIC) method was used to calculate the strain field of a sample experiencing large deformations during a tensile test. Two codes were used for the calculation (the iCorrelate-vision 2D code and GOM software). Three calculation options were applied; the spatial correlation of images was evaluated, as was the incremental correlation of successive images. A comparison of the results is given in the article. Mainly, the decorrelation problem arises due to the poor matching between the correlated subsets in the case of large strains. Poor matching can result from large changes in the speckle pattern. Updating the reference image can minimize these changes. When using a fixed reference image, the correlation algorithm will not be able to track significant changes in the deformed target subset, which will lead to a failure in the matching between the reference and deformed subsets, even when using correlation algorithms with very high efficiency