Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.29975 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916061467639808 |
|---|---|
| author | Nellikunnummel, Nisar Barbour, Andi Wiegart, Lutz Konstantinova, Tatiana DeGennaro, Anthony |
| author_facet | Nellikunnummel, Nisar Barbour, Andi Wiegart, Lutz Konstantinova, Tatiana DeGennaro, Anthony |
| contents | 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 of the dataset and reduce overfitting. The FC-DAE successfully recovers intricate dynamical features in low signal-to-noise conditions while maintaining structural fidelity. To assess reconstruction reliability, we employ quantitative metrics to evaluate structural fidelity and identify potential model-induced bias. Our results demonstrate that the FC-DAE provides robust denoising performance with high computational efficiency, enabling recovery of XPCS dynamics under photon-limited and low-dose measurement conditions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_29975 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Fully Convolutional Approach to Denoising Structural Dynamics Data from X-Ray Photon Correlation Spectroscopy Nellikunnummel, Nisar Barbour, Andi Wiegart, Lutz Konstantinova, Tatiana DeGennaro, Anthony Machine Learning Signal Processing 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 of the dataset and reduce overfitting. The FC-DAE successfully recovers intricate dynamical features in low signal-to-noise conditions while maintaining structural fidelity. To assess reconstruction reliability, we employ quantitative metrics to evaluate structural fidelity and identify potential model-induced bias. Our results demonstrate that the FC-DAE provides robust denoising performance with high computational efficiency, enabling recovery of XPCS dynamics under photon-limited and low-dose measurement conditions. |
| title | A Fully Convolutional Approach to Denoising Structural Dynamics Data from X-Ray Photon Correlation Spectroscopy |
| topic | Machine Learning Signal Processing |
| url | https://arxiv.org/abs/2605.29975 |