On the use of adaptive refinement in isogeometric digital image correlation

A novel global digital image correlation method was developed using adaptive refinement of isogeometric shape functions. Non-uniform rational B-spline shape functions are used because of their flexibility and versatility, which enable them to capture a wide range of kinematics. The goal of this work was to explore the full potential of isogeometric shape functions for digital image correlation (DIC). This is reached by combining a global DIC method with an adaptive refinement algorithm: adaptive isogeometric GDIC. The shape functions are automatically adjusted to be able to describe the kinematics of the sought displacement field with an optimized number of degrees of freedom. This results in an accurate method without the need of making problem-specific choices regarding the structure of the shape functions, which makes the method less user input dependent than regular global DIC methods, while keeping the number of degrees of freedom limited to realize optimum regularization of the ill-posed DIC problem. The method's accuracy is demonstrated by a virtual experiment with a predefined, highly localized displacement field. Real experiments with a complex sample geometry demonstrate the effectiveness in practice.