Universal image segmentation for optical identification of 2D materials

Abstract Machine learning methods are changing the way data is analyzed. One of the most powerful and widespread applications of these techniques is in image segmentation wherein disparate objects of a digital image are partitioned and classified. Here we present an image segmentation program incorp...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Randy M. Sterbentz, Kristine L. Haley, Joshua O. Island
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/bc46e0476ced4b9ca84a9176522342e8
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:bc46e0476ced4b9ca84a9176522342e8
record_format dspace
spelling oai:doaj.org-article:bc46e0476ced4b9ca84a9176522342e82021-12-02T15:54:14ZUniversal image segmentation for optical identification of 2D materials10.1038/s41598-021-85159-92045-2322https://doaj.org/article/bc46e0476ced4b9ca84a9176522342e82021-03-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-85159-9https://doaj.org/toc/2045-2322Abstract Machine learning methods are changing the way data is analyzed. One of the most powerful and widespread applications of these techniques is in image segmentation wherein disparate objects of a digital image are partitioned and classified. Here we present an image segmentation program incorporating a series of unsupervised clustering algorithms for the automatic thickness identification of two-dimensional materials from digital optical microscopy images. The program identifies mono- and few-layer flakes of a variety of materials on both opaque and transparent substrates with a pixel accuracy of roughly 95%. Contrasting with previous attempts, application generality is achieved through preservation and analysis of all three digital color channels and Gaussian mixture model fits to arbitrarily shaped data clusters. Our results provide a facile implementation of data clustering for the universal, automatic identification of two-dimensional materials exfoliated onto any substrate.Randy M. SterbentzKristine L. HaleyJoshua O. IslandNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-8 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Randy M. Sterbentz
Kristine L. Haley
Joshua O. Island
Universal image segmentation for optical identification of 2D materials
description Abstract Machine learning methods are changing the way data is analyzed. One of the most powerful and widespread applications of these techniques is in image segmentation wherein disparate objects of a digital image are partitioned and classified. Here we present an image segmentation program incorporating a series of unsupervised clustering algorithms for the automatic thickness identification of two-dimensional materials from digital optical microscopy images. The program identifies mono- and few-layer flakes of a variety of materials on both opaque and transparent substrates with a pixel accuracy of roughly 95%. Contrasting with previous attempts, application generality is achieved through preservation and analysis of all three digital color channels and Gaussian mixture model fits to arbitrarily shaped data clusters. Our results provide a facile implementation of data clustering for the universal, automatic identification of two-dimensional materials exfoliated onto any substrate.
format article
author Randy M. Sterbentz
Kristine L. Haley
Joshua O. Island
author_facet Randy M. Sterbentz
Kristine L. Haley
Joshua O. Island
author_sort Randy M. Sterbentz
title Universal image segmentation for optical identification of 2D materials
title_short Universal image segmentation for optical identification of 2D materials
title_full Universal image segmentation for optical identification of 2D materials
title_fullStr Universal image segmentation for optical identification of 2D materials
title_full_unstemmed Universal image segmentation for optical identification of 2D materials
title_sort universal image segmentation for optical identification of 2d materials
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/bc46e0476ced4b9ca84a9176522342e8
work_keys_str_mv AT randymsterbentz universalimagesegmentationforopticalidentificationof2dmaterials
AT kristinelhaley universalimagesegmentationforopticalidentificationof2dmaterials
AT joshuaoisland universalimagesegmentationforopticalidentificationof2dmaterials
_version_ 1718385472801079296