Digital image correlation is a branch of photogrammetry aimed at assessing sub-pixel accurate deformations of a material or industrial solution (e.g. fire extinguisher) by analysing the deformation of the pattern applied to it. This approach can be done in 2D or 3D, for which 1 to 2 cameras will be needed. The main difference between the two lies in whether or not out-of-plane deformations can be assessed. Basically, digital image correlation involves the following steps:
- Camera/camera calibration
- Use of an appropriately sized black and white mottling (to achieve maximum contrast between areas).
- Analysis of the displacement experienced by a region (called a subset) between the image at time i and the image at time The deformation of this region is performed by comparing the two regions, first by cross-correlation (approximate position) and then by refinement through a non-linear Gauss-Newton optimisation.
- Analysis of the rest of the regions using the same approach.
As a result, it is possible to evaluate the mechanical properties of materials (e.g. Young’s modulus or Poisson’s ratio), which allows the validation of complex numerical simulations using the finite element method.