Dictionary-learning-based image deblurring for improving image performance in x-ray nondestructive testing

This study investigated a dictionary-learning (DL)-based image deblurring method for improving image performance in x-ray nondestructive testing. DL is a representation learning theory that aims to find a sparse representation of the input signal in the form of a linear combination of basic elements as well as those basic elements themselves. In this study, a DL-based algorithm was implemented, and a computational simulation and experiment were then performed to evaluate the algorithm's effectiveness for image deblurring. The hardware system used in the experiment consisted of an x-ray tube with a focal spot size of 0.6 mm and a flat-panel detector with a pixel size of 100 $\mu$m². X-ray images of several electronic components were acquired at x-ray tube conditions of 80 kV${}_p$ and 1.25 mAs. The image characteristics of the deblurred images generated by the DL-based algorithm were quantitatively evaluated in terms of intensity profile, universal-quality index, and noise power spectrum. Our results indicate that our DL-based image deblurring method effectively improves image performance in x-ray nondestructive testing.