Automated Contrast Calculation and Non-Linear Fitting for Defect Quantification in Infrared Thermography

Abstract

Infrared thermography is a powerful non-destructive testing technique that can detect and locate near-surface defects in materials based on their thermal contrast. However, accurately quantifying the size and depth of defects from thermal images remains a challenging task, particularly when the heating pattern is non-uniform. In this paper, we present a novel approach that combines automated contrast calculation with non-linear fitting to enable accurate defect quantification in infrared thermography. Our method involves automated detection of the defect-free region based on the analysis of apparent thermal effusivity to automatically calculate the thermal contrast between the defect and the surrounding area. This contrast is then used as input to a non-linear fitting algorithm that models the thermal response of the material and estimates the size and depth of the defect. We evaluate the performance of our method on simulated and experimental data and show that it outperforms traditional methods based on absolute contrast and phase analysis. Our approach provides a reliable and automated way to quantify defects in infrared thermography and has the potential to improve the efficiency and accuracy of non-destructive testing in a wide range of applications