A Quantitative Comparison Among Different Algorithms for Defects Detection on Aluminum with the Pulsed Thermography Technique

Abstract

Pulsed thermography is commonly used as a non-destructive technique for evaluating defects within materials and components. In the last few years, many algorithms have been developed with the aim to detect defects and different methods have been used for detecting their size and depth. However, only few works in the literature reported a comparison among the different algorithms in terms of the number of detected defects, the time spent in testing and analysis, and the quantitative evaluation of size and depth. In this work, starting from a pulsed thermographic test carried out on an aluminum specimen with twenty flat bottom holes of known nominal size and depth, different algorithms have been used with the aim to obtain a comparison among them in terms of signal to background contrast (SBC) and number of detected defects by analyzing different time intervals. Moreover, the correlation between SBC and the aspect ratio of the defects has been investigated. The algorithms used have been: Pulsed Phase Thermography (PPT), Slope, Correlation Coefficient (R2), Thermal Signal Reconstruction (TSR) and Principal Component Thermography (PCT). The results showed the advantages, disadvantages, and sensitivity of the various thermographic algorithms.