Digital image correlation with improved efficiency by pixel selection.

With the increase in digital image correlation (DIC) applications, the computational efficiency of DIC is becoming increasingly important. In previous studies, real-time DIC was realized with a relatively small subset. However, a small subset does not always include sufficient gray gradient information. In this paper, a pixel selection strategy is proposed to improve the computational efficiency of DIC further, allowing a real-time deformation measurement with a large subset. Within the subset, zero weight is assigned to unreliable pixels as a way of pursuing maximum efficiency. The modulus of the local intensity gradient vector of each pixel in the reference image is used as the criterion for reliability. Numerical and real experiments conducted to validate the feasibility and effectiveness of the strategy showed that the computational speed of DIC could be improved about 2 times.