Critical comparative assessment of digital image correlation vs contact extensometry: ensuring the fidelity and depth of mechanical properties determination

• Cross-platform DIC gives better than 0.2 % strain agreement with contact extensometry. • Virtual extensometer position has negligible effect on DIC strain accuracy. • DIC enables full-field strain mapping and early detection of localization. • Optimized subset & filtering minimize DIC noise to ±0.2 % strain. • Applicable to LPBF miniature specimens under constrained testing conditions. This study presents a comprehensive experimental comparison between the Digital Image Correlation (DIC) technique and a mechanical contact extensometer for determining the mechanical properties of AISI 321 (Russian grade 12Kh18N10T) austenitic stainless steel under uniaxial tensile loading. The investigation focuses on two novel aspects: (i) the influence of virtual extensometer placement within the DIC region of interest and (ii) cross-platform validation using both Vic-3D (commercial) and Ncorr (open-source) software. Tensile tests were performed on six dog-bone specimens using an Instron 5969 universal testing machine (maximum load: 50 kN). Both strain-measurement techniques were applied simultaneously. The results revealed strong quantitative agreement: Young’s modulus ranged from 194.7 to 196.2 GPa, yield strength from 245 to 260 MPa, and ultimate tensile strength from 574 to 584 MPa. The relative deviation between DIC and contact extensometry remained within 2 % for modulus and yield strength and below 1 % for ultimate strength. No statistically significant influence of virtual extensometer positioning was observed within the homogeneous gauge zone. Cross-platform comparison confirmed the consistency of DIC data between Vic-3D and Ncorr, with an inter-software deviation of less than 1 %. DIC strain curves exhibited characteristic waviness due to numerical differentiation; however, Gaussian smoothing and optimized subset parameters (31 × 31 px) effectively reduced noise to ±0.2 % strain. This work advances the field by establishing a quantitative, cross-validated methodology for evaluating DIC accuracy relative to contact extensometry and by demonstrating its applicability to miniature or non-standard specimens, particularly those produced via additive manufacturing (LPBF). The findings confirm DIC as a reliable, non-contact alternative for precise strain measurement and material characterization under static loading.