Dynamic feature sampling method analysis for the detection of microcrack on uncoated aluminum alloy surface by joint scanning laser thermography

Laser thermography is a cutting-edge non-destructive testing method with remarkable advantages such as contactless operation, swift detection, and exceptional sensitivity. In this study, an innovative thermal signal sampling method is introduced for the detection of microcracks on uncoated aluminum alloy surface through joint scanning laser thermography. Initially, numerical simulation was conducted to analyze the thermal response characteristics of the uncoated surface during inspection. Subsequently, the dynamic minimum value features of the surface temperature distribution were extracted within a large region of interest, and the optimal sampling region was determined by evaluating the signal-to-noise ratio. The temperature curves of the sampling results showed a clear peak-valley bipolar feature at the microcrack locations. The correlation coefficient of one-dimensional continuous wavelet transform was employed to extract the singularity of crack features, and the singularity points were clearly defined to locate the microcracks spatial location. The experimental results showed that our proposed detection method could quickly and automatically locate and characterize microcracks on the surface of uncoated aluminum alloy with good stability and robustness. This method provides a feasible solution for the application of laser thermography in industry, thus improving the quality and safety performance of products.