Saved in:
Bibliographic Details
Main Authors: Degnan-Morgenstern, Samuel, Cohen, Alexander E., Gopal, Rajeev, Gober, Megan, Nelson, George J., Bai, Peng, Bazant, Martin Z.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.27667
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917052887859200
author Degnan-Morgenstern, Samuel
Cohen, Alexander E.
Gopal, Rajeev
Gober, Megan
Nelson, George J.
Bai, Peng
Bazant, Martin Z.
author_facet Degnan-Morgenstern, Samuel
Cohen, Alexander E.
Gopal, Rajeev
Gober, Megan
Nelson, George J.
Bai, Peng
Bazant, Martin Z.
contents Operando microscopy provides direct insight into the dynamic chemical and physical processes that govern functional materials, yet measurement noise limits the effective resolution and undermines quantitative analysis. Here, we present a general framework for integrating unsupervised deep learning-based denoising into quantitative microscopy workflows across modalities and length scales. Using simulated data, we demonstrate that deep denoising preserves physical fidelity, introduces minimal bias, and reduces uncertainty in model learning with partial differential equation (PDE)-constrained optimization. Applied to experiments, denoising reveals nanoscale chemical and structural heterogeneity in scanning transmission X-ray microscopy (STXM) of lithium iron phosphate (LFP), enables automated particle segmentation and phase classification in optical microscopy of graphite electrodes, and reduces noise-induced variability by nearly 80% in neutron radiography to resolve heterogeneous lithium transport. Collectively, these results establish deep denoising as a powerful, modality-agnostic enhancement that advances quantitative operando imaging and extends the reach of previously noise-limited techniques.
format Preprint
id arxiv_https___arxiv_org_abs_2510_27667
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Deep learning denoising unlocks quantitative insights in operando materials microscopy
Degnan-Morgenstern, Samuel
Cohen, Alexander E.
Gopal, Rajeev
Gober, Megan
Nelson, George J.
Bai, Peng
Bazant, Martin Z.
Computer Vision and Pattern Recognition
Materials Science
Operando microscopy provides direct insight into the dynamic chemical and physical processes that govern functional materials, yet measurement noise limits the effective resolution and undermines quantitative analysis. Here, we present a general framework for integrating unsupervised deep learning-based denoising into quantitative microscopy workflows across modalities and length scales. Using simulated data, we demonstrate that deep denoising preserves physical fidelity, introduces minimal bias, and reduces uncertainty in model learning with partial differential equation (PDE)-constrained optimization. Applied to experiments, denoising reveals nanoscale chemical and structural heterogeneity in scanning transmission X-ray microscopy (STXM) of lithium iron phosphate (LFP), enables automated particle segmentation and phase classification in optical microscopy of graphite electrodes, and reduces noise-induced variability by nearly 80% in neutron radiography to resolve heterogeneous lithium transport. Collectively, these results establish deep denoising as a powerful, modality-agnostic enhancement that advances quantitative operando imaging and extends the reach of previously noise-limited techniques.
title Deep learning denoising unlocks quantitative insights in operando materials microscopy
topic Computer Vision and Pattern Recognition
Materials Science
url https://arxiv.org/abs/2510.27667