Coupled digital image correlation and infrared thermography applied to curved specimens

The availability of experimental data is essential for constitutive model calibration and validation of numerical results. Advances in optical imaging techniques enable the measurement of surface temperatures via infrared thermography and surface deformations using digital image correlation. Thus, coupling both measurement methods to obtain temperature and deformation data at material points is of significant interest. However, it is a disadvantage that both methods employ different measurement principles. Infrared thermography captures temperature data at fixed pixels, operating in an Eulerian framework, whereas digital image correlation tracks specific points during deformation, thus operating in a Lagrangian setting. The relation between both datasets is straightforward for flat specimens. However, specimens with curved surfaces such as those in tension-torsion tests, require a more sophisticated approach utilizing curvilinear coordinates.

The projection of material points into the two-dimensional thermography images is described by a camera model. Calibrating the camera model and determining the transformation between both measurement systems is done by means of a reference object. Although parallel projection (orthographic camera) has been successfully used in previous studies, it has limitations in projection accuracy. Consequently, central projection (pinhole camera) is investigated as an alternative, demonstrating improved projection accuracy compared to parallel projection. The overall methodology for coupling deformation and temperature measurements is explained. Therein, we utilize global interpolation for both experimental datasets with radial basis functions, allowing for strain analysis and computation of temperature gradients in curved surfaces.