Error Analysis of the Digital Image Correlation Method

Abstract

The optical method Digital Image Correlation (DIC) has been extensively studied due its capability to measure the entire displacement field over a body surface. This particular feature makes this measurement method appropriated to measure heterogeneous displacement fields in large strains. Then, it can be applied for specimens used in standard tests or even for in vivo applications, like biological tissues. In the last years, many works about the performance of different DIC algorithms were published. These studies demonstrated the excellent accuracy and precision of the optical method; however almost all have been focused on the measurement errors for low order displacement fields, like translation and constant strain, and only a few works have been studied the error related to high-order displacement fields. Since this method shown great applicability for heterogeneous displacement fields founded in large strains, this work investigates the displacement measure accuracy for different possible sets of the DIC algorithm using numerically simulated images for zero, first and second order displacement fields in order to consider only errors related to the algorithm. The results shown that the accuracy may vary in many orders of magnitude from the choice of different setups, where the relation between the subset size and the speckle granule size is quite important.