Improving Spatial Resolution and Correcting Surface Topography in Portable Macro-XRF Imaging Devices
Ze He,Zhu An,Ning Huang,Peng Wang,Jingxi Yuan,Hai Wang,Jiamei Wang,Jie He
DOI: https://doi.org/10.1109/tns.2024.3443124
IF: 1.703
2024-09-21
IEEE Transactions on Nuclear Science
Abstract:In macro X-ray fluorescence (MA-XRF) imaging devices, a larger beam size leads to worse spatial resolution, and the surface topography can produce shadow patterns in the image that hinder its analysis. In this study, a portable MA-XRF imaging device is constructed, and the performance of the device is measured. Three image reconstruction algorithms are introduced, including compressive sensing, L0-regularized intensity and gradient prior (L0RIGP), and blind deconvolution (BD), to compare the image reconstruction of the USAF 1951 resolution test target. The results show that the BD algorithm outperformed the other two methods, achieving up to a 35% improvement in spatial resolution, a 24.2% enhancement in image signal-to-noise ratio, and an optimal spatial resolution of 0.36 mm (1.39 lp/mm). An intensity theoretical calculation model of XRF based on the sample surface topography is established. The copper map of a "5 Jiao" coin is corrected and the mean deviation decreased by 5.7%. Finally, a piece of lapis lazuli is analyzed, and the quality of the element maps of K, Ca, and Sr is improved by the surface topography correction and image reconstruction, with spatial resolutions increasing by 36.1%, 34.4%, and 33.5%, respectively. The research in this article demonstrates that the MA-XRF imaging device based on spatial resolution optimization and surface topography correction algorithms has promising applications.
engineering, electrical & electronic,nuclear science & technology