Heat haze removal in digital image correlation using a grayscale weighted average algorithm

Digital image correlation has emerged as a vital quantitative non-destructive evaluation (QNDE) tool for quantifying full-field deformation in high-temperature material characterisation. Nevertheless, measurement accuracy faces critical challenges from heat haze induced image distortions. This study presents a novel Grayscale Weighted Averaging algorithm based on Zero-Mean Normalised Cross-Correlation to mitigate these artefacts. Numerical simulations and experimental validation show that the proposed GWA method improves the peak signal-to-noise ratio (PSNR) by 5% compared with conventional grayscale averaging. Simulated tensile-strain analysis shows a 28.6% reduction in relative error and a 50% decrease in displacement standard deviation. Experiments conducted in a simulated aero-engine operating environment at 1423 K and high temperature furnace at 573 K confirm the robustness of the algorithm, achieving a 49.1% artefact suppression while maintaining displacement field regularity at 1423 K while the U-displacement coefficient of dispersion decreases by 23.2%. This optical optimisation strategy advances high-temperature DIC metrology by simultaneously enhancing image quality and reducing measurement uncertainty, particularly valuable for high-temperature digital image correlation.