A novel method to compensate systematic errors due to undermatched shape functions in digital image correlation

Abstract

Undermatched shape functions have always been the primary source of systematic errors for non-uniform deformation measurement in digital image correlation, especially for large curvature displacement. All the other errors arising from grey-level interpolation, subset size, and speckle pattern are overwhelmed by the systematic errors due to undermatched shape functions. The effect of the first-order shape functions on polynomials is analyzed, and based on this analysis, a novel and easy to implement displacement post-processing algorithm is proposed to compensate the systematic errors. The direct digital image correlation displacement result is processed by multiple filters with the same window size as the subset, and the resulting data is employed to recover the actual displacement field. Theoretically, the proposed method is capable of recovering the original continuous displacement function of any order with no bias. The numerical results of elementary function and simulated images show that the proposed algorithm is effective in correcting the displacement systematic bias and improving the subpixel accuracy of digital image correlation. Finally, the algorithm is applied to the displacement measurement of shear band.