Artifact segmentation using the U-Net architecture for powder X-ray diffraction images

Modern synchrotron X-ray facilities generate X-ray diffraction (XRD) image data at rates that far exceed the capacity of manual analysis, yet critical processing steps such as artifact removal rely on human intervention. Rule-based algorithms in standard beamline software cannot reliably distinguish harmful artifacts like single-crystal diffraction spots from desirable features like preferred orientation, which leads to incorrectly processed data. We demonstrate that a U-Net convolutional neural network trained on expert-labeled experimental data can help automate artifact segmentation in time-resolved XRD measurements of battery materials. To address overfitting from redundant time-series frames, we introduce a mutual information-based pruning algorithm that selects maximally diverse training images. Our optimized model achieves an 85.1% true positive rate on a fully held-out battery dataset while reducing false positives by 34% compared with GSAS-II. Critically, the model preserves preferred orientation features that GSAS-II incorrectly removes, preventing systematic underestimation of phase intensities in downstream analysis.

# Nuclear and particle physics. Atomic energy. Radioactivity # Crystallography # artifact segmentation # computer vision # machine learning # x-ray imaging

Albert Vong, Howard Yanxon, Eric Roberts, et al. (2026). "Artifact segmentation using the U-Net architecture for powder X-ray diffraction images". Journal of Synchrotron Radiation. https://doi.org/10.1107/s160057752600754x