Structure-aware guided filtering for a ring artifact correction in synchrotron x-ray microtomography

Abstract

Synchrotron-based x-ray microtomography (S-µCT) is a powerful non-invasive three-dimensional (3D) imaging technique used for visualizing the internal structure of objects with micron-scale spatial resolution. However, in practical applications, ring artifacts often occur in S-µCT, which significantly degrades image quality and hinders interpretation. In this study, we propose a ring artifact correction method based on guided image filtering (GIF). The method first extracts structural prior from the input S-µCT images and then uses it as the guidance image to correct the ring artifacts. Finally, GIF with a self-guidance image is employed to further enhance image quality. Extensive comparisons and analyses on simulations and real data experiments demonstrate that the proposed method is capable of effectively correcting ring artifacts, accompanied by low-dose noise suppression and sparse-view artifact reduction. These findings suggest that the proposed method has great potential to promote the wider applications of S-µCT in the future.