Strain field measurements in the tensile tests using digital image correlation

Abstract

Developments in digital image correlation (DIC) to measure strain fields are very useful for tensile tests because it is a non-contact metrology technique. Many studies use the sub-pixel registration algorithms of DIC to increase measurement accuracy but there are few studies reported on measurement errors. Then, the aim of this paper is to develop the sub-pixel registration algorithm based on Newton-Raphson iteration that is evaluated for its performances by using computer-generated speckle images. The subset sizes for evaluation are small (21´21 pixels), medium (31´31 pixels) and large (41´41 pixels). It is shown that the systematic errors and the random errors of small subsets are greater than other subset sizes. Not only subset sizes but also the speckle sizes play an important role for errors. Next, real tensile tests were performed to investigate the measurement results of DIC compared with a video extensometer. The smoothing technique is used for noise reduction in the strain fields before plotting stress-strain curves. Finally, the results of real tensile tests indicate that DIC can be capable to measure strain fields and Young’s modulus.