Predictive visualization of fiber laser cutting topography via deep learning with image inpainting

Predictive visualization of fiber laser cutting topography via deep learning with image inpainting

Laser cutting is a fast, precise, and noncontact processing technique widely applied throughout industry. However, parameter specific defects can be formed while cutting, negatively impacting the cut quality. While light-matter interactions are highly nonlinear and are, therefore, challenging to mod...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of laser applications 2023-08, Vol.35 (3)
Hauptverfasser: Courtier, Alexander F., Praeger, Matthew, Grant-Jacob, James A., Codemard, Christophe, Harrison, Paul, Zervas, Michalis, Mills, Ben
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue 3
container_start_page
container_title Journal of laser applications
container_volume 35
creator Courtier, Alexander F.
Praeger, Matthew
Grant-Jacob, James A.
Codemard, Christophe
Harrison, Paul
Zervas, Michalis
Mills, Ben
description Laser cutting is a fast, precise, and noncontact processing technique widely applied throughout industry. However, parameter specific defects can be formed while cutting, negatively impacting the cut quality. While light-matter interactions are highly nonlinear and are, therefore, challenging to model analytically, deep learning offers the capability of modeling these interactions directly from data. Here, we show that deep learning can be used to scale up visual predictions for parameter specific defects produced in cutting as well as for predicting defects for parameters not measured experimentally. Furthermore, visual predictions can be used to model the relationship between laser cutting defects and laser cutting parameters.
doi_str_mv 10.2351/7.0000957
format Article
fullrecord <record><control><sourceid>scitation_cross</sourceid><recordid>TN_cdi_crossref_primary_10_2351_7_0000957</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>jla</sourcerecordid><originalsourceid>FETCH-LOGICAL-c334t-3e08d8cdf2f77fc7ba80f2cb22eaf581175aa849a74c81844cfa195d27d905ca3</originalsourceid><addsrcrecordid>eNp9kE1Lw0AQhhdRsFYP_oO9KqTuV9zNUYpWoaAHBW9hsh_tSkzC7jZSf72JLXoQnMPMwDzzMvMidE7JjPGcXskZGaLI5QGa0IKrjHIlD4eeCJZxcf16jE5ifCOESi7FBFVPwRqvk-8t7n3cQO0_Ifm2wa3Dzlc24BrikPUmJd-scGq7dhWgW28HHrCxtsO1hdCMww-f1ti_w8pi33Tgm3HlFB05qKM929cperm7fZ7fZ8vHxcP8ZplpzkXKuCXKKG0cc1I6LStQxDFdMWbB5YpSmQMoUYAUWlElhHZAi9wwaQqSa-BTdLHT1aGNMVhXdmG4JWxLSsrRnFKWe3MG9nLHRu3T978_cN-GX7DsjPsP_qv8BXE9dYs</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Predictive visualization of fiber laser cutting topography via deep learning with image inpainting</title><source>AIP Journals Complete</source><creator>Courtier, Alexander F. ; Praeger, Matthew ; Grant-Jacob, James A. ; Codemard, Christophe ; Harrison, Paul ; Zervas, Michalis ; Mills, Ben</creator><creatorcontrib>Courtier, Alexander F. ; Praeger, Matthew ; Grant-Jacob, James A. ; Codemard, Christophe ; Harrison, Paul ; Zervas, Michalis ; Mills, Ben</creatorcontrib><description>Laser cutting is a fast, precise, and noncontact processing technique widely applied throughout industry. However, parameter specific defects can be formed while cutting, negatively impacting the cut quality. While light-matter interactions are highly nonlinear and are, therefore, challenging to model analytically, deep learning offers the capability of modeling these interactions directly from data. Here, we show that deep learning can be used to scale up visual predictions for parameter specific defects produced in cutting as well as for predicting defects for parameters not measured experimentally. Furthermore, visual predictions can be used to model the relationship between laser cutting defects and laser cutting parameters.</description><identifier>ISSN: 1042-346X</identifier><identifier>EISSN: 1938-1387</identifier><identifier>DOI: 10.2351/7.0000957</identifier><identifier>CODEN: JLAPEN</identifier><language>eng</language><ispartof>Journal of laser applications, 2023-08, Vol.35 (3)</ispartof><rights>Author(s)</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c334t-3e08d8cdf2f77fc7ba80f2cb22eaf581175aa849a74c81844cfa195d27d905ca3</citedby><cites>FETCH-LOGICAL-c334t-3e08d8cdf2f77fc7ba80f2cb22eaf581175aa849a74c81844cfa195d27d905ca3</cites><orcidid>0000-0002-1784-1012 ; 0000-0002-0651-4059 ; 0009-0004-6390-1137 ; 0000-0002-4270-4247 ; 0000-0002-5814-6155 ; 0000-0001-9189-8263 ; 0000-0003-1943-4055</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://pubs.aip.org/jla/article-lookup/doi/10.2351/7.0000957$$EHTML$$P50$$Gscitation$$Hfree_for_read</linktohtml><link.rule.ids>314,777,781,791,4498,27905,27906,76133</link.rule.ids></links><search><creatorcontrib>Courtier, Alexander F.</creatorcontrib><creatorcontrib>Praeger, Matthew</creatorcontrib><creatorcontrib>Grant-Jacob, James A.</creatorcontrib><creatorcontrib>Codemard, Christophe</creatorcontrib><creatorcontrib>Harrison, Paul</creatorcontrib><creatorcontrib>Zervas, Michalis</creatorcontrib><creatorcontrib>Mills, Ben</creatorcontrib><title>Predictive visualization of fiber laser cutting topography via deep learning with image inpainting</title><title>Journal of laser applications</title><description>Laser cutting is a fast, precise, and noncontact processing technique widely applied throughout industry. However, parameter specific defects can be formed while cutting, negatively impacting the cut quality. While light-matter interactions are highly nonlinear and are, therefore, challenging to model analytically, deep learning offers the capability of modeling these interactions directly from data. Here, we show that deep learning can be used to scale up visual predictions for parameter specific defects produced in cutting as well as for predicting defects for parameters not measured experimentally. Furthermore, visual predictions can be used to model the relationship between laser cutting defects and laser cutting parameters.</description><issn>1042-346X</issn><issn>1938-1387</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNp9kE1Lw0AQhhdRsFYP_oO9KqTuV9zNUYpWoaAHBW9hsh_tSkzC7jZSf72JLXoQnMPMwDzzMvMidE7JjPGcXskZGaLI5QGa0IKrjHIlD4eeCJZxcf16jE5ifCOESi7FBFVPwRqvk-8t7n3cQO0_Ifm2wa3Dzlc24BrikPUmJd-scGq7dhWgW28HHrCxtsO1hdCMww-f1ti_w8pi33Tgm3HlFB05qKM929cperm7fZ7fZ8vHxcP8ZplpzkXKuCXKKG0cc1I6LStQxDFdMWbB5YpSmQMoUYAUWlElhHZAi9wwaQqSa-BTdLHT1aGNMVhXdmG4JWxLSsrRnFKWe3MG9nLHRu3T978_cN-GX7DsjPsP_qv8BXE9dYs</recordid><startdate>202308</startdate><enddate>202308</enddate><creator>Courtier, Alexander F.</creator><creator>Praeger, Matthew</creator><creator>Grant-Jacob, James A.</creator><creator>Codemard, Christophe</creator><creator>Harrison, Paul</creator><creator>Zervas, Michalis</creator><creator>Mills, Ben</creator><scope>AJDQP</scope><scope>AAYXX</scope><scope>CITATION</scope><orcidid>https://orcid.org/0000-0002-1784-1012</orcidid><orcidid>https://orcid.org/0000-0002-0651-4059</orcidid><orcidid>https://orcid.org/0009-0004-6390-1137</orcidid><orcidid>https://orcid.org/0000-0002-4270-4247</orcidid><orcidid>https://orcid.org/0000-0002-5814-6155</orcidid><orcidid>https://orcid.org/0000-0001-9189-8263</orcidid><orcidid>https://orcid.org/0000-0003-1943-4055</orcidid></search><sort><creationdate>202308</creationdate><title>Predictive visualization of fiber laser cutting topography via deep learning with image inpainting</title><author>Courtier, Alexander F. ; Praeger, Matthew ; Grant-Jacob, James A. ; Codemard, Christophe ; Harrison, Paul ; Zervas, Michalis ; Mills, Ben</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c334t-3e08d8cdf2f77fc7ba80f2cb22eaf581175aa849a74c81844cfa195d27d905ca3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Courtier, Alexander F.</creatorcontrib><creatorcontrib>Praeger, Matthew</creatorcontrib><creatorcontrib>Grant-Jacob, James A.</creatorcontrib><creatorcontrib>Codemard, Christophe</creatorcontrib><creatorcontrib>Harrison, Paul</creatorcontrib><creatorcontrib>Zervas, Michalis</creatorcontrib><creatorcontrib>Mills, Ben</creatorcontrib><collection>AIP Open Access Journals</collection><collection>CrossRef</collection><jtitle>Journal of laser applications</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Courtier, Alexander F.</au><au>Praeger, Matthew</au><au>Grant-Jacob, James A.</au><au>Codemard, Christophe</au><au>Harrison, Paul</au><au>Zervas, Michalis</au><au>Mills, Ben</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Predictive visualization of fiber laser cutting topography via deep learning with image inpainting</atitle><jtitle>Journal of laser applications</jtitle><date>2023-08</date><risdate>2023</risdate><volume>35</volume><issue>3</issue><issn>1042-346X</issn><eissn>1938-1387</eissn><coden>JLAPEN</coden><abstract>Laser cutting is a fast, precise, and noncontact processing technique widely applied throughout industry. However, parameter specific defects can be formed while cutting, negatively impacting the cut quality. While light-matter interactions are highly nonlinear and are, therefore, challenging to model analytically, deep learning offers the capability of modeling these interactions directly from data. Here, we show that deep learning can be used to scale up visual predictions for parameter specific defects produced in cutting as well as for predicting defects for parameters not measured experimentally. Furthermore, visual predictions can be used to model the relationship between laser cutting defects and laser cutting parameters.</abstract><doi>10.2351/7.0000957</doi><tpages>17</tpages><orcidid>https://orcid.org/0000-0002-1784-1012</orcidid><orcidid>https://orcid.org/0000-0002-0651-4059</orcidid><orcidid>https://orcid.org/0009-0004-6390-1137</orcidid><orcidid>https://orcid.org/0000-0002-4270-4247</orcidid><orcidid>https://orcid.org/0000-0002-5814-6155</orcidid><orcidid>https://orcid.org/0000-0001-9189-8263</orcidid><orcidid>https://orcid.org/0000-0003-1943-4055</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1042-346X
ispartof Journal of laser applications, 2023-08, Vol.35 (3)
issn 1042-346X
1938-1387
language eng
recordid cdi_crossref_primary_10_2351_7_0000957
source AIP Journals Complete
title Predictive visualization of fiber laser cutting topography via deep learning with image inpainting
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-18T14%3A33%3A57IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-scitation_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Predictive%20visualization%20of%20fiber%20laser%20cutting%20topography%20via%20deep%20learning%20with%20image%20inpainting&rft.jtitle=Journal%20of%20laser%20applications&rft.au=Courtier,%20Alexander%20F.&rft.date=2023-08&rft.volume=35&rft.issue=3&rft.issn=1042-346X&rft.eissn=1938-1387&rft.coden=JLAPEN&rft_id=info:doi/10.2351/7.0000957&rft_dat=%3Cscitation_cross%3Ejla%3C/scitation_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true