Exploring Image Quality Improvements in High-Speed Dual Threshold Photon-Counting Micro-CT

Abstract High-speed X-ray computed tomography (CT) of batteries in-line or at-line is a promising technique to obtain quality-relevant insights leading to an optimized production process and detection of defective batteries. By using a high-power micro-focus X-ray source and a photon-counting detector, CT scans can be obtained within seconds. Here we explore utilizing the simultaneous readout of multiple images at different energy-discriminating thresholds and recombining them to improve the quality of the reconstructed volumes to optimize different quality parameters relevant to battery inspection. Using a-priori knowledge, threshold optimization is performed. Evaluating the combined volumes shows that there is an ideal threshold, or combination of two thresholds, depending on what matric used to optimize contrast between specific feature of a specific sample. Further, the contrast of the jelly roll compared to the rest of the battery can also be improved by combining two different thresholds. The experiments highlight the importance of threshold optimization and the potential gain of combining two simultaneous acquisitions using different energy thresholds for fast CT scans with limited photon statistics.