DIC-based identication of the constitutive parameters of an elastoplastic law is addressed both from a general viewpoint, and applied to the particular case of dog-bone sample made of commercially pure titanium and subjected to tensile loading. A two-step procedure (Digital Image Correlation DIC followed by weighted Finite Element Method Updating FEMU) is rst presented. These two steps can be merged into a single-step procedure (i.e., Integrated-DIC or I-DIC). In both cases, the elastoplastic computations are performed with a commercial code (i.e., non-intrusive identication). When the suited weighting of FEMU is taken into account, which is based on DIC-processed image noise, both I-DIC and FEMU methods provide similar results. It is shown that the addressed experimental case requires the use of static (load) information to get precise estimates of the sought parameters.

It is proposed to use digital image correlation (DIC) to identify parameters governing crack propagation of commercially pure titanium. To achieve this goal, crack tip location, stress intensity factor, T -stress and plastic zone size are sought. Most of the DIC approaches are based upon local analyses of displacements, and their subsequent projection onto a set of mechanically relevant fields. It is proposed to perform these two sequential steps in a unique (and integrated) way, and to compare the results with a global approach to DIC with subsequent post-processing. A priori performances of two global approaches are compared, and a propagation law is identified form the series of raw images of a fatigue test on commercially pure titanium with the integrated approach that yields better results.