_version_ 1866911777130807296
author Oppliger, Jens
Denner, M. Michael
Küspert, Julia
Frison, Ruggero
Wang, Qisi
Morawietz, Alexander
Ivashko, Oleh
Dippel, Ann-Christin
von Zimmermann, Martin
Biało, Izabela
Martinelli, Leonardo
Fauqué, Benoît
Choi, Jaewon
Garcia-Fernandez, Mirian
Zhou, Ke-Jin
Christensen, Niels B.
Kurosawa, Tohru
Momono, Naoki
Oda, Migaku
Natterer, Fabian D.
Fischer, Mark H.
Neupert, Titus
Chang, Johan
author_facet Oppliger, Jens
Denner, M. Michael
Küspert, Julia
Frison, Ruggero
Wang, Qisi
Morawietz, Alexander
Ivashko, Oleh
Dippel, Ann-Christin
von Zimmermann, Martin
Biało, Izabela
Martinelli, Leonardo
Fauqué, Benoît
Choi, Jaewon
Garcia-Fernandez, Mirian
Zhou, Ke-Jin
Christensen, Niels B.
Kurosawa, Tohru
Momono, Naoki
Oda, Migaku
Natterer, Fabian D.
Fischer, Mark H.
Neupert, Titus
Chang, Johan
contents Removal or cancellation of noise has wide-spread applications for imaging and acoustics. In every-day-life applications, denoising may even include generative aspects, which are unfaithful to the ground truth. For scientific use, however, denoising must reproduce the ground truth accurately. Here, we show how data can be denoised via a deep convolutional neural network such that weak signals appear with quantitative accuracy. In particular, we study X-ray diffraction on crystalline materials. We demonstrate that weak signals stemming from charge ordering, insignificant in the noisy data, become visible and accurate in the denoised data. This success is enabled by supervised training of a deep neural network with pairs of measured low- and high-noise data. We demonstrate that using artificial noise does not yield such quantitatively accurate results. Our approach thus illustrates a practical strategy for noise filtering that can be applied to challenging acquisition problems.
format Preprint
id arxiv_https___arxiv_org_abs_2209_09247
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Weak-signal extraction enabled by deep-neural-network denoising of diffraction data
Oppliger, Jens
Denner, M. Michael
Küspert, Julia
Frison, Ruggero
Wang, Qisi
Morawietz, Alexander
Ivashko, Oleh
Dippel, Ann-Christin
von Zimmermann, Martin
Biało, Izabela
Martinelli, Leonardo
Fauqué, Benoît
Choi, Jaewon
Garcia-Fernandez, Mirian
Zhou, Ke-Jin
Christensen, Niels B.
Kurosawa, Tohru
Momono, Naoki
Oda, Migaku
Natterer, Fabian D.
Fischer, Mark H.
Neupert, Titus
Chang, Johan
Image and Video Processing
Strongly Correlated Electrons
Superconductivity
Machine Learning
Removal or cancellation of noise has wide-spread applications for imaging and acoustics. In every-day-life applications, denoising may even include generative aspects, which are unfaithful to the ground truth. For scientific use, however, denoising must reproduce the ground truth accurately. Here, we show how data can be denoised via a deep convolutional neural network such that weak signals appear with quantitative accuracy. In particular, we study X-ray diffraction on crystalline materials. We demonstrate that weak signals stemming from charge ordering, insignificant in the noisy data, become visible and accurate in the denoised data. This success is enabled by supervised training of a deep neural network with pairs of measured low- and high-noise data. We demonstrate that using artificial noise does not yield such quantitatively accurate results. Our approach thus illustrates a practical strategy for noise filtering that can be applied to challenging acquisition problems.
title Weak-signal extraction enabled by deep-neural-network denoising of diffraction data
topic Image and Video Processing
Strongly Correlated Electrons
Superconductivity
Machine Learning
url https://arxiv.org/abs/2209.09247