Automatised quality assessment in additive layer manufacturing using layer-by-layer surface measurements and deep learning

Léopold Le Roux; Chao Liu; Ze Ji; Pierre Kerfriden; Daniel Gage; Felix Feyer; Carolin Körner; Samuel Bigot

doi:10.1016/j.procir.2021.03.050

Automatised quality assessment in additive layer manufacturing using layer-by-layer surface measurements and deep learning

Léopold Le Roux
, Chao Liu
, Ze Ji
, Pierre Kerfriden
, Daniel Gage
, Felix Feyer
, Carolin Körner
, Samuel Bigot^*

^*Corresponding author for this work

Mechanical, Biomedical & Design Engineering

Cardiff Metropolitan University
Mines ParisTech
Friedrich-Alexander-Universität Erlangen-Nürnberg

Research output: Contribution to journal › Conference article › peer-review

14 Citations (Scopus)

12 Downloads (Pure)

Abstract

Additive manufacturing (AM) has gained high research interests in the past but comes with some drawbacks, such as the difficulty to do in-situ quality monitoring. In this paper, deep learning is used on electron-optical images taken during the Electron Beam Melting (EBM) process to classify the quality of AM layers to achieve automatized quality assessment. A comparative study of several mainstream Convolutional Neural Networks to classify the images has been conducted. The classification accuracy is up to 95 %, which demonstrates the great potential to support in-process layer quality control of EBM.And the error analysis has shown that some human misclassification were correctly classified by the Convolutional Neural Networks.

Original language	English
Pages (from-to)	342-347
Number of pages	6
Journal	Procedia CIRP
Volume	99
Early online date	3 May 2021
DOIs	https://doi.org/10.1016/j.procir.2021.03.050
Publication status	Published - 2021
Event	14th CIRP Conference on Intelligent Computation in Manufacturing Engineering, CIRP ICME 2020 - Naples, Italy Duration: 15 Jul 2020 → 17 Jul 2020

Bibliographical note

© 2021 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0)
Peer-review under responsibility of the scientific committee of the 14th CIRP Conference on Intelligent Computation in Manufacturing Engineering 15-17 July 2020

Funding Information:
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement n°820774.

Funding

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement n°820774.

Keywords

Additive Manufacturing
Artificial Intelligence
Image recognition
Quality control
Transfert learning

Access to Document

10.1016/j.procir.2021.03.050

Automatised quality assessment in additive layer manufacturing
© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the 14th CIRP Conference on Intelligent Computation in Manufacturing Engineering 15-17 July 2020
Final published version, 953 KBLicence: CC BY-NC-ND 3.0

Cite this

@article{9765166e4c8d479f8dd8b02f45e1a372,

title = "Automatised quality assessment in additive layer manufacturing using layer-by-layer surface measurements and deep learning",

abstract = "Additive manufacturing (AM) has gained high research interests in the past but comes with some drawbacks, such as the difficulty to do in-situ quality monitoring. In this paper, deep learning is used on electron-optical images taken during the Electron Beam Melting (EBM) process to classify the quality of AM layers to achieve automatized quality assessment. A comparative study of several mainstream Convolutional Neural Networks to classify the images has been conducted. The classification accuracy is up to 95 \%, which demonstrates the great potential to support in-process layer quality control of EBM.And the error analysis has shown that some human misclassification were correctly classified by the Convolutional Neural Networks.",

keywords = "Additive Manufacturing, Artificial Intelligence, Image recognition, Quality control, Transfert learning",

author = "Roux, \{L{\'e}opold Le\} and Chao Liu and Ze Ji and Pierre Kerfriden and Daniel Gage and Felix Feyer and Carolin K{\"o}rner and Samuel Bigot",

note = "{\textcopyright} 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the 14th CIRP Conference on Intelligent Computation in Manufacturing Engineering 15-17 July 2020 Funding Information: This project has received funding from the European Union{\textquoteright}s Horizon 2020 research and innovation programme under grant agreement n°820774. ; 14th CIRP Conference on Intelligent Computation in Manufacturing Engineering, CIRP ICME 2020 ; Conference date: 15-07-2020 Through 17-07-2020",

year = "2021",

doi = "10.1016/j.procir.2021.03.050",

language = "English",

volume = "99",

pages = "342--347",

journal = "Procedia CIRP",

issn = "2212-8271",

publisher = "Elsevier",

}

TY - JOUR

T1 - Automatised quality assessment in additive layer manufacturing using layer-by-layer surface measurements and deep learning

AU - Roux, Léopold Le

AU - Liu, Chao

AU - Ji, Ze

AU - Kerfriden, Pierre

AU - Gage, Daniel

AU - Feyer, Felix

AU - Körner, Carolin

AU - Bigot, Samuel

N1 - © 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the 14th CIRP Conference on Intelligent Computation in Manufacturing Engineering 15-17 July 2020 Funding Information: This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement n°820774.

PY - 2021

Y1 - 2021

N2 - Additive manufacturing (AM) has gained high research interests in the past but comes with some drawbacks, such as the difficulty to do in-situ quality monitoring. In this paper, deep learning is used on electron-optical images taken during the Electron Beam Melting (EBM) process to classify the quality of AM layers to achieve automatized quality assessment. A comparative study of several mainstream Convolutional Neural Networks to classify the images has been conducted. The classification accuracy is up to 95 %, which demonstrates the great potential to support in-process layer quality control of EBM.And the error analysis has shown that some human misclassification were correctly classified by the Convolutional Neural Networks.

AB - Additive manufacturing (AM) has gained high research interests in the past but comes with some drawbacks, such as the difficulty to do in-situ quality monitoring. In this paper, deep learning is used on electron-optical images taken during the Electron Beam Melting (EBM) process to classify the quality of AM layers to achieve automatized quality assessment. A comparative study of several mainstream Convolutional Neural Networks to classify the images has been conducted. The classification accuracy is up to 95 %, which demonstrates the great potential to support in-process layer quality control of EBM.And the error analysis has shown that some human misclassification were correctly classified by the Convolutional Neural Networks.

KW - Additive Manufacturing

KW - Artificial Intelligence

KW - Image recognition

KW - Quality control

KW - Transfert learning

UR - https://www.scopus.com/pages/publications/85106439688

UR - https://www.sciencedirect.com/science/article/pii/S2212827121003267?via%3Dihub

U2 - 10.1016/j.procir.2021.03.050

DO - 10.1016/j.procir.2021.03.050

M3 - Conference article

AN - SCOPUS:85106439688

SN - 2212-8271

VL - 99

SP - 342

EP - 347

JO - Procedia CIRP

JF - Procedia CIRP

T2 - 14th CIRP Conference on Intelligent Computation in Manufacturing Engineering, CIRP ICME 2020

Y2 - 15 July 2020 through 17 July 2020

ER -

Automatised quality assessment in additive layer manufacturing using layer-by-layer surface measurements and deep learning

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this