A Fully Convolutional Approach to Denoising Structural Dynamics Data from X-Ray Photon Correlation Spectroscopy
arXiv:2605.29975v1 Announce Type: new Abstract: We present a fully convolutional denoising autoencoder (FC-DAE) for denoising two-time intensity-intensity correlation functions ($C_2$) in X-ray photon correlation spectroscopy (XPCS). Unlike conventional denoising autoencoders that are typically restricted to fixed input sizes, the FC-DAE accepts inputs of arbitrary dimensions while preserving correlation structures across diverse dynamical regimes. The model is trained using experimentally derived $C_2$ data collected at NSLS-II beamlines, with data augmentation applied to expand the diversity
This research paper is published as part of the ongoing academic advancement in AI and scientific data analysis, reflecting continuous development in specialized applications.
It represents incremental progress in using AI for scientific data processing, potentially improving the efficiency and accuracy of X-ray photon correlation spectroscopy.
The development proposes a method for denoising specific scientific data, offering a technical refinement within a niche scientific domain rather than a broad systemic change.
- · Materials science researchers
- · Synchrotron facilities
Improved data quality for experimental measurements using XPCS.
Potentially faster analysis workflows for scientists studying structural dynamics.
Slightly accelerated discovery of new material properties through more reliable data interpretation.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG