A Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials

X-Ray Computed Tomography (XCT) techniques have evolved to a point that high-resolution data can be acquired so fast that classic segmentation methods are prohibitively cumbersome, demanding automated data pipelines capable of dealing with non-trivial 3D images. Meanwhile, deep learning has demonstr...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	João P. C. Bertoldo, Etienne Decencière , David Ryckelynck , Henry Proudhon
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	deep learning U-net modular network architecture semantic segmentation 3D X-ray computed tomography (XCT) composite material Technology T
Acceso en línea:	https://doaj.org/article/679dbf87b1ff4732838daf11f3be2271
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:679dbf87b1ff4732838daf11f3be2271
record_format	dspace
spelling	oai:doaj.org-article:679dbf87b1ff4732838daf11f3be22712021-11-30T23:09:05ZA Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials2296-801610.3389/fmats.2021.761229https://doaj.org/article/679dbf87b1ff4732838daf11f3be22712021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fmats.2021.761229/fullhttps://doaj.org/toc/2296-8016X-Ray Computed Tomography (XCT) techniques have evolved to a point that high-resolution data can be acquired so fast that classic segmentation methods are prohibitively cumbersome, demanding automated data pipelines capable of dealing with non-trivial 3D images. Meanwhile, deep learning has demonstrated success in many image processing tasks, including materials science applications, showing a promising alternative for a human-free segmentation pipeline. However, the rapidly increasing number of available architectures can be a serious drag to the wide adoption of this type of models by the end user. In this paper a modular interpretation of U-Net (Modular U-Net) is proposed with a parametrized architecture that can be easily tuned to optimize it. As an example, the model is trained to segment 3D tomography images of a three-phased glass fiber-reinforced Polyamide 66. We compare 2D and 3D versions of our model, finding that the former is slightly better than the latter. We observe that human-comparable results can be achievied even with only 13 annotated slices and using a shallow U-Net yields better results than a deeper one. As a consequence, neural networks show indeed a promising venue to automate XCT data processing pipelines needing no human, adhoc intervention.João P. C. BertoldoEtienne Decencière David Ryckelynck Henry ProudhonFrontiers Media S.A.articledeep learningU-netmodular network architecturesemantic segmentation3D X-ray computed tomography (XCT)composite materialTechnologyTENFrontiers in Materials, Vol 8 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	deep learning U-net modular network architecture semantic segmentation 3D X-ray computed tomography (XCT) composite material Technology T
spellingShingle	deep learning U-net modular network architecture semantic segmentation 3D X-ray computed tomography (XCT) composite material Technology T João P. C. Bertoldo Etienne Decencière David Ryckelynck Henry Proudhon A Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials
description	X-Ray Computed Tomography (XCT) techniques have evolved to a point that high-resolution data can be acquired so fast that classic segmentation methods are prohibitively cumbersome, demanding automated data pipelines capable of dealing with non-trivial 3D images. Meanwhile, deep learning has demonstrated success in many image processing tasks, including materials science applications, showing a promising alternative for a human-free segmentation pipeline. However, the rapidly increasing number of available architectures can be a serious drag to the wide adoption of this type of models by the end user. In this paper a modular interpretation of U-Net (Modular U-Net) is proposed with a parametrized architecture that can be easily tuned to optimize it. As an example, the model is trained to segment 3D tomography images of a three-phased glass fiber-reinforced Polyamide 66. We compare 2D and 3D versions of our model, finding that the former is slightly better than the latter. We observe that human-comparable results can be achievied even with only 13 annotated slices and using a shallow U-Net yields better results than a deeper one. As a consequence, neural networks show indeed a promising venue to automate XCT data processing pipelines needing no human, adhoc intervention.
format	article
author	João P. C. Bertoldo Etienne Decencière David Ryckelynck Henry Proudhon
author_facet	João P. C. Bertoldo Etienne Decencière David Ryckelynck Henry Proudhon
author_sort	João P. C. Bertoldo
title	A Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials
title_short	A Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials
title_full	A Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials
title_fullStr	A Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials
title_full_unstemmed	A Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials
title_sort	modular u-net for automated segmentation of x-ray tomography images in composite materials
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/679dbf87b1ff4732838daf11f3be2271
work_keys_str_mv	AT joaopcbertoldo amodularunetforautomatedsegmentationofxraytomographyimagesincompositematerials AT etiennedecenciere amodularunetforautomatedsegmentationofxraytomographyimagesincompositematerials AT davidryckelynck amodularunetforautomatedsegmentationofxraytomographyimagesincompositematerials AT henryproudhon amodularunetforautomatedsegmentationofxraytomographyimagesincompositematerials AT joaopcbertoldo modularunetforautomatedsegmentationofxraytomographyimagesincompositematerials AT etiennedecenciere modularunetforautomatedsegmentationofxraytomographyimagesincompositematerials AT davidryckelynck modularunetforautomatedsegmentationofxraytomographyimagesincompositematerials AT henryproudhon modularunetforautomatedsegmentationofxraytomographyimagesincompositematerials
_version_	1718406205421912064

A Modular U-Net for Automated Segmentation of X-Ray Tomography Images in Composite Materials

Ejemplares similares