Comparative study of deep learning algorithms for atomic force microscopy image denoising

Hoichan Jung; Giwoong Han; Seong Jun Jung; Sung Won Han

doi:10.1016/j.micron.2022.103332

Comparative study of deep learning algorithms for atomic force microscopy image denoising

Hoichan Jung, Giwoong Han, Seong Jun Jung, Sung Won Han

Research output: Contribution to journal › Article › peer-review

4 Citations (Scopus)

Abstract

Atomic force microscopy (AFM) enables direct visualisation of surface topography at the nanoscale. However, post-processing is generally required to obtain accurate, precise, and reliable AFM images owing to the presence of image artefacts. In this study, we compared and analysed state-of-the-art deep learning models, namely MPRNet, HINet, Uformer, and Restormer, with respect to denoising AFM images containing four types of noise. Specifically, these algorithms’ denoising performance and inference time on AFM images were compared with those of conventional methods and previous studies. Through a comparative analysis, we found that the most efficient and the most effective models were Restormer and HINet, respectively. The code, models, and data used in this work are available at https://github.com/hoichanjung/AFM_Image_Denoising.

Original language	English
Article number	103332
Journal	Micron
Volume	161
DOIs	https://doi.org/10.1016/j.micron.2022.103332
Publication status	Published - 2022 Oct

Bibliographical note

Funding Information:
This research was supported by Brain Korea 21 FOUR . This research was also supported by Korea Institute for Advancement of Technology (KIAT) grant funded by the Korea Government (MOTIE) ( P0008691 , The Competency Development Program for Industry Specialist).

Publisher Copyright:
© 2022 Elsevier Ltd

Keywords

Atomic force microscopy image
Deep neural network
Image denoising
Image restoration

ASJC Scopus subject areas

Structural Biology
General Materials Science
General Physics and Astronomy
Cell Biology

Access to Document

10.1016/j.micron.2022.103332

Cite this

@article{362a7f2e9492405b9ca1b462450fb372,

title = "Comparative study of deep learning algorithms for atomic force microscopy image denoising",

abstract = "Atomic force microscopy (AFM) enables direct visualisation of surface topography at the nanoscale. However, post-processing is generally required to obtain accurate, precise, and reliable AFM images owing to the presence of image artefacts. In this study, we compared and analysed state-of-the-art deep learning models, namely MPRNet, HINet, Uformer, and Restormer, with respect to denoising AFM images containing four types of noise. Specifically, these algorithms{\textquoteright} denoising performance and inference time on AFM images were compared with those of conventional methods and previous studies. Through a comparative analysis, we found that the most efficient and the most effective models were Restormer and HINet, respectively. The code, models, and data used in this work are available at https://github.com/hoichanjung/AFM_Image_Denoising.",

keywords = "Atomic force microscopy image, Deep neural network, Image denoising, Image restoration",

author = "Hoichan Jung and Giwoong Han and Jung, {Seong Jun} and Han, {Sung Won}",

note = "Funding Information: This research was supported by Brain Korea 21 FOUR . This research was also supported by Korea Institute for Advancement of Technology (KIAT) grant funded by the Korea Government (MOTIE) ( P0008691 , The Competency Development Program for Industry Specialist). Publisher Copyright: {\textcopyright} 2022 Elsevier Ltd",

year = "2022",

month = oct,

doi = "10.1016/j.micron.2022.103332",

language = "English",

volume = "161",

journal = "Micron",

issn = "0968-4328",

publisher = "Elsevier Ltd",

}

TY - JOUR

T1 - Comparative study of deep learning algorithms for atomic force microscopy image denoising

AU - Jung, Hoichan

AU - Han, Giwoong

AU - Jung, Seong Jun

AU - Han, Sung Won

N1 - Funding Information: This research was supported by Brain Korea 21 FOUR . This research was also supported by Korea Institute for Advancement of Technology (KIAT) grant funded by the Korea Government (MOTIE) ( P0008691 , The Competency Development Program for Industry Specialist). Publisher Copyright: © 2022 Elsevier Ltd

PY - 2022/10

Y1 - 2022/10

N2 - Atomic force microscopy (AFM) enables direct visualisation of surface topography at the nanoscale. However, post-processing is generally required to obtain accurate, precise, and reliable AFM images owing to the presence of image artefacts. In this study, we compared and analysed state-of-the-art deep learning models, namely MPRNet, HINet, Uformer, and Restormer, with respect to denoising AFM images containing four types of noise. Specifically, these algorithms’ denoising performance and inference time on AFM images were compared with those of conventional methods and previous studies. Through a comparative analysis, we found that the most efficient and the most effective models were Restormer and HINet, respectively. The code, models, and data used in this work are available at https://github.com/hoichanjung/AFM_Image_Denoising.

AB - Atomic force microscopy (AFM) enables direct visualisation of surface topography at the nanoscale. However, post-processing is generally required to obtain accurate, precise, and reliable AFM images owing to the presence of image artefacts. In this study, we compared and analysed state-of-the-art deep learning models, namely MPRNet, HINet, Uformer, and Restormer, with respect to denoising AFM images containing four types of noise. Specifically, these algorithms’ denoising performance and inference time on AFM images were compared with those of conventional methods and previous studies. Through a comparative analysis, we found that the most efficient and the most effective models were Restormer and HINet, respectively. The code, models, and data used in this work are available at https://github.com/hoichanjung/AFM_Image_Denoising.

KW - Atomic force microscopy image

KW - Deep neural network

KW - Image denoising

KW - Image restoration

UR - http://www.scopus.com/inward/record.url?scp=85136107320&partnerID=8YFLogxK

U2 - 10.1016/j.micron.2022.103332

DO - 10.1016/j.micron.2022.103332

M3 - Article

C2 - 35952420

AN - SCOPUS:85136107320

SN - 0968-4328

VL - 161

JO - Micron

JF - Micron

M1 - 103332

ER -

Comparative study of deep learning algorithms for atomic force microscopy image denoising

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this