Search papers, labs, and topics across Lattice.
This paper introduces a fully convolutional denoising autoencoder (FC-DAE) for denoising two-time intensity-intensity correlation functions ($C_2$) obtained from X-ray photon correlation spectroscopy (XPCS). The FC-DAE overcomes the input size limitations of traditional denoising autoencoders, allowing it to handle variable-sized XPCS data while preserving correlation structures. Trained on experimental data from NSLS-II with data augmentation, the FC-DAE effectively recovers dynamic features in low signal-to-noise conditions, offering a computationally efficient approach to XPCS data analysis.
Unlock hidden dynamics in noisy X-ray experiments: a fully convolutional autoencoder now efficiently denoises variable-sized correlation functions, even under photon-limited conditions.
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.
Amazon Science