Standard

Nanoparticle recognition on scanning probe microscopy images using computer vision and deep learning. / Okunev, Alexey G.; Mashukov, Mikhail Yu; Nartova, Anna V. и др.

в: Nanomaterials, Том 10, № 7, 1285, 30.06.2020, стр. 1-16.

Результаты исследований: Научные публикации в периодических изданияхстатьяРецензирование

Harvard

APA

Vancouver

Okunev AG, Mashukov MY, Nartova AV, Matveev AV. Nanoparticle recognition on scanning probe microscopy images using computer vision and deep learning. Nanomaterials. 2020 июнь 30;10(7):1-16. 1285. doi: 10.3390/nano10071285

Author

Okunev, Alexey G. ; Mashukov, Mikhail Yu ; Nartova, Anna V. и др. / Nanoparticle recognition on scanning probe microscopy images using computer vision and deep learning. в: Nanomaterials. 2020 ; Том 10, № 7. стр. 1-16.

BibTeX

@article{a7a020660ae7428a90fcb17e7ec7282d,
title = "Nanoparticle recognition on scanning probe microscopy images using computer vision and deep learning",
abstract = "Identifying, counting and measuring particles is an important component of many research studies. Images with particles are usually processed by hand using a software ruler. Automated processing, based on conventional image processing methods (edge detection, segmentation, etc.) are not universal, can only be used on good-quality images and need to set a number of parameters empirically. In this paper, we present results from the application of deep learning to automated recognition of metal nanoparticles deposited on highly oriented pyrolytic graphite on images obtained by scanning tunneling microscopy (STM). We used the Cascade Mask-RCNN neural network. Training was performed on a dataset containing 23 STM images with 5157 nanoparticles. Three images containing 695 nanoparticles were used for verification. As a result, the trained neural network recognized nanoparticles in the verification set with 0.93 precision and 0.78 recall. Predicted contour refining with 2D Gaussian function was a proposed option. The accuracies for mean particle size calculated from predicted contours compared with ground truth were in the range of 0.87–0.99. The results were compared with outcomes from other generally available software, based on conventional image processing methods. The advantages of deep learning methods for automatic particle recognition were clearly demonstrated. We developed a free open-access web service “ParticlesNN” based on the trained neural network, which can be used by any researcher in the world.",
keywords = "Deep neural networks, Particle recognition, Particles, Scanning tunneling microscopy",
author = "Okunev, {Alexey G.} and Mashukov, {Mikhail Yu} and Nartova, {Anna V.} and Matveev, {Andrey V.}",
note = "Publisher Copyright: {\textcopyright} 2020 by the authors. Licensee MDPI, Basel, Switzerland. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.",
year = "2020",
month = jun,
day = "30",
doi = "10.3390/nano10071285",
language = "English",
volume = "10",
pages = "1--16",
journal = "Nanomaterials",
issn = "2079-4991",
publisher = "MDPI AG",
number = "7",

}

RIS

TY - JOUR

T1 - Nanoparticle recognition on scanning probe microscopy images using computer vision and deep learning

AU - Okunev, Alexey G.

AU - Mashukov, Mikhail Yu

AU - Nartova, Anna V.

AU - Matveev, Andrey V.

N1 - Publisher Copyright: © 2020 by the authors. Licensee MDPI, Basel, Switzerland. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.

PY - 2020/6/30

Y1 - 2020/6/30

N2 - Identifying, counting and measuring particles is an important component of many research studies. Images with particles are usually processed by hand using a software ruler. Automated processing, based on conventional image processing methods (edge detection, segmentation, etc.) are not universal, can only be used on good-quality images and need to set a number of parameters empirically. In this paper, we present results from the application of deep learning to automated recognition of metal nanoparticles deposited on highly oriented pyrolytic graphite on images obtained by scanning tunneling microscopy (STM). We used the Cascade Mask-RCNN neural network. Training was performed on a dataset containing 23 STM images with 5157 nanoparticles. Three images containing 695 nanoparticles were used for verification. As a result, the trained neural network recognized nanoparticles in the verification set with 0.93 precision and 0.78 recall. Predicted contour refining with 2D Gaussian function was a proposed option. The accuracies for mean particle size calculated from predicted contours compared with ground truth were in the range of 0.87–0.99. The results were compared with outcomes from other generally available software, based on conventional image processing methods. The advantages of deep learning methods for automatic particle recognition were clearly demonstrated. We developed a free open-access web service “ParticlesNN” based on the trained neural network, which can be used by any researcher in the world.

AB - Identifying, counting and measuring particles is an important component of many research studies. Images with particles are usually processed by hand using a software ruler. Automated processing, based on conventional image processing methods (edge detection, segmentation, etc.) are not universal, can only be used on good-quality images and need to set a number of parameters empirically. In this paper, we present results from the application of deep learning to automated recognition of metal nanoparticles deposited on highly oriented pyrolytic graphite on images obtained by scanning tunneling microscopy (STM). We used the Cascade Mask-RCNN neural network. Training was performed on a dataset containing 23 STM images with 5157 nanoparticles. Three images containing 695 nanoparticles were used for verification. As a result, the trained neural network recognized nanoparticles in the verification set with 0.93 precision and 0.78 recall. Predicted contour refining with 2D Gaussian function was a proposed option. The accuracies for mean particle size calculated from predicted contours compared with ground truth were in the range of 0.87–0.99. The results were compared with outcomes from other generally available software, based on conventional image processing methods. The advantages of deep learning methods for automatic particle recognition were clearly demonstrated. We developed a free open-access web service “ParticlesNN” based on the trained neural network, which can be used by any researcher in the world.

KW - Deep neural networks

KW - Particle recognition

KW - Particles

KW - Scanning tunneling microscopy

UR - http://www.scopus.com/inward/record.url?scp=85087371596&partnerID=8YFLogxK

U2 - 10.3390/nano10071285

DO - 10.3390/nano10071285

M3 - Article

C2 - 32629955

AN - SCOPUS:85087371596

VL - 10

SP - 1

EP - 16

JO - Nanomaterials

JF - Nanomaterials

SN - 2079-4991

IS - 7

M1 - 1285

ER -

ID: 24722411