Deep-learning versus greyscale segmentation of voids in X-ray computed tomography images of filament-wound composites

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Standard

Deep-learning versus greyscale segmentation of voids in X-ray computed tomography images of filament-wound composites. / Upadhyay, Shailee; George Smith, Abraham; Vandepitte, Dirk; Lomov, Stepan V.; Swolfs, Yentl; Mehdikhani, Mahoor.

I: Composites Part A: Applied Science and Manufacturing, Bind 177, 107937, 2024.

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Harvard

Upadhyay, S, George Smith, A, Vandepitte, D, Lomov, SV, Swolfs, Y & Mehdikhani, M 2024, 'Deep-learning versus greyscale segmentation of voids in X-ray computed tomography images of filament-wound composites', Composites Part A: Applied Science and Manufacturing, bind 177, 107937. https://doi.org/10.1016/j.compositesa.2023.107937

APA

Upadhyay, S., George Smith, A., Vandepitte, D., Lomov, S. V., Swolfs, Y., & Mehdikhani, M. (2024). Deep-learning versus greyscale segmentation of voids in X-ray computed tomography images of filament-wound composites. Composites Part A: Applied Science and Manufacturing, 177, [107937]. https://doi.org/10.1016/j.compositesa.2023.107937

Vancouver

Upadhyay S, George Smith A, Vandepitte D, Lomov SV, Swolfs Y, Mehdikhani M. Deep-learning versus greyscale segmentation of voids in X-ray computed tomography images of filament-wound composites. Composites Part A: Applied Science and Manufacturing. 2024;177. 107937. https://doi.org/10.1016/j.compositesa.2023.107937

Author

Upadhyay, Shailee ; George Smith, Abraham ; Vandepitte, Dirk ; Lomov, Stepan V. ; Swolfs, Yentl ; Mehdikhani, Mahoor. / Deep-learning versus greyscale segmentation of voids in X-ray computed tomography images of filament-wound composites. I: Composites Part A: Applied Science and Manufacturing. 2024 ; Bind 177.

Bibtex

@article{070bc2f105fc45ad9650e8ddb3d04dba,
title = "Deep-learning versus greyscale segmentation of voids in X-ray computed tomography images of filament-wound composites",
abstract = "Filament-wound composites (FWC) are prone to high void contents, with large and complex-shape voids. It is critical to characterise these voids accurately to understand their effect on part strength. The characterization depends on the accuracy of the analysis technique, for example X-ray computed tomography and the subsequent void segmentation. This paper compares conventional greyscale thresholding to deep-learning (DL) based segmentation. The processing steps for both techniques are discussed. The greyscale thresholding contains segmentation errors due to the simple one-parameter algorithm and the pre-processing operations required for segmentation. This reduces the accuracy of void characterisation. The DL-based segmentation is found to be more accurate for characterisation of void size, shape, and location. The processing-time and system requirements are discussed, helping to determine the suitable segmentation technique based on desired results.",
keywords = "A. Carbon fibre, B. Porosity, D. CT analysis, E. Filament winding",
author = "Shailee Upadhyay and {George Smith}, Abraham and Dirk Vandepitte and Lomov, {Stepan V.} and Yentl Swolfs and Mahoor Mehdikhani",
note = "Publisher Copyright: {\textcopyright} 2023 Elsevier Ltd",
year = "2024",
doi = "10.1016/j.compositesa.2023.107937",
language = "English",
volume = "177",
journal = "Composites Part A: Applied Science and Manufacturing",
issn = "1359-835X",
publisher = "Elsevier",

}

RIS

TY - JOUR

T1 - Deep-learning versus greyscale segmentation of voids in X-ray computed tomography images of filament-wound composites

AU - Upadhyay, Shailee

AU - George Smith, Abraham

AU - Vandepitte, Dirk

AU - Lomov, Stepan V.

AU - Swolfs, Yentl

AU - Mehdikhani, Mahoor

N1 - Publisher Copyright: © 2023 Elsevier Ltd

PY - 2024

Y1 - 2024

N2 - Filament-wound composites (FWC) are prone to high void contents, with large and complex-shape voids. It is critical to characterise these voids accurately to understand their effect on part strength. The characterization depends on the accuracy of the analysis technique, for example X-ray computed tomography and the subsequent void segmentation. This paper compares conventional greyscale thresholding to deep-learning (DL) based segmentation. The processing steps for both techniques are discussed. The greyscale thresholding contains segmentation errors due to the simple one-parameter algorithm and the pre-processing operations required for segmentation. This reduces the accuracy of void characterisation. The DL-based segmentation is found to be more accurate for characterisation of void size, shape, and location. The processing-time and system requirements are discussed, helping to determine the suitable segmentation technique based on desired results.

AB - Filament-wound composites (FWC) are prone to high void contents, with large and complex-shape voids. It is critical to characterise these voids accurately to understand their effect on part strength. The characterization depends on the accuracy of the analysis technique, for example X-ray computed tomography and the subsequent void segmentation. This paper compares conventional greyscale thresholding to deep-learning (DL) based segmentation. The processing steps for both techniques are discussed. The greyscale thresholding contains segmentation errors due to the simple one-parameter algorithm and the pre-processing operations required for segmentation. This reduces the accuracy of void characterisation. The DL-based segmentation is found to be more accurate for characterisation of void size, shape, and location. The processing-time and system requirements are discussed, helping to determine the suitable segmentation technique based on desired results.

KW - A. Carbon fibre

KW - B. Porosity

KW - D. CT analysis

KW - E. Filament winding

U2 - 10.1016/j.compositesa.2023.107937

DO - 10.1016/j.compositesa.2023.107937

M3 - Journal article

AN - SCOPUS:85180375861

VL - 177

JO - Composites Part A: Applied Science and Manufacturing

JF - Composites Part A: Applied Science and Manufacturing

SN - 1359-835X

M1 - 107937

ER -

ID: 378176718