A Deep Learning Approach to Identify Local Structures in Atomic-Resolution Transmission Electron Microscopy Images

Research output: Contribution to journalJournal article – Annual report year: 2018Researchpeer-review

Standard

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{55ee860301bd49ba85f2d19a14c2452e,
title = "A Deep Learning Approach to Identify Local Structures in Atomic-Resolution Transmission Electron Microscopy Images",
abstract = "Recording atomic-resolution transmission electron microscopy (TEM) images isbecoming increasingly routine. A new bottleneck is then analyzing thisinformation, which often involves time-consuming manual structuralidentification. We have developed a deep learning-based algorithm forrecognition of the local structure in TEM images, which is stable to microscopeparameters and noise. The neural network is trained entirely from simulationbut is capable of making reliable predictions on experimental images. We applythe method to single sheets of defected graphene, and to metallic nanoparticleson an oxide support.",
author = "Jacob Madsen and Pei Liu and Jens Kling and Wagner, {Jakob Birkedal} and Hansen, {Thomas Willum} and Ole Winther and Jakob Schi{\o}tz",
year = "2018",
doi = "10.1002/adts.201800037",
language = "English",
volume = "1",
journal = "Advanced Theory and Simulations",
issn = "2513-0390",
number = "8",

}

RIS

TY - JOUR

T1 - A Deep Learning Approach to Identify Local Structures in Atomic-Resolution Transmission Electron Microscopy Images

AU - Madsen, Jacob

AU - Liu, Pei

AU - Kling, Jens

AU - Wagner, Jakob Birkedal

AU - Hansen, Thomas Willum

AU - Winther, Ole

AU - Schiøtz, Jakob

PY - 2018

Y1 - 2018

N2 - Recording atomic-resolution transmission electron microscopy (TEM) images isbecoming increasingly routine. A new bottleneck is then analyzing thisinformation, which often involves time-consuming manual structuralidentification. We have developed a deep learning-based algorithm forrecognition of the local structure in TEM images, which is stable to microscopeparameters and noise. The neural network is trained entirely from simulationbut is capable of making reliable predictions on experimental images. We applythe method to single sheets of defected graphene, and to metallic nanoparticleson an oxide support.

AB - Recording atomic-resolution transmission electron microscopy (TEM) images isbecoming increasingly routine. A new bottleneck is then analyzing thisinformation, which often involves time-consuming manual structuralidentification. We have developed a deep learning-based algorithm forrecognition of the local structure in TEM images, which is stable to microscopeparameters and noise. The neural network is trained entirely from simulationbut is capable of making reliable predictions on experimental images. We applythe method to single sheets of defected graphene, and to metallic nanoparticleson an oxide support.

U2 - 10.1002/adts.201800037

DO - 10.1002/adts.201800037

M3 - Journal article

VL - 1

JO - Advanced Theory and Simulations

JF - Advanced Theory and Simulations

SN - 2513-0390

IS - 8

M1 - 1800037

ER -