An Improved Lightweight YOLOv11 Algorithm for Weld Surface Defect Detection

An Improved Lightweight YOLOv11 Algorithm for Weld Surface Defect Detection | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article An Improved Lightweight YOLOv11 Algorithm for Weld Surface Defect Detection Runmei Zhang, Chenfei Pan, Zihua Chen, Jingwei Fan, Zhong Chen, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7808482/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 28 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted 12 You are reading this latest preprint version Abstract Industrial welding often exhibits some essential problems, such as unclear defect characteristics and complex backgroundinformation. However, the existing defect detection models have relatively high costs and may be weak in weld defect detection.To address the problem, this paper proposes an improved lightweight YOLOv11 model for welding surface defect detection,called YOLO-Air. First, the model integrates the feature extraction module with the convolutional module to enhance featureextraction capabilities and optimize computational resource utilization. Second, the GSConv and VOV-GSCSP modules areembedded in the neck network to reduce feature redundancy in both spatial and channel dimensions, and then lower thecomputational load. Third, a lightweight detection head is designed as part of the detection network to further reduce modelcomplexity. Finally, we compare our proposed YOLO-Air model to the baseline model via the Welding Defect Test-V2 andNEU-DET dataset. Experimental results show that the proposed model achieves superior weld surface detection effects,especially the metric mAP50 is +1.3%, the parameter number is reduced by 17.3%, and the computational load is reduced by 31.7% Physical sciences/Engineering Physical sciences/Mathematics and computing Deep learning Defect detection YOLOv11 PConv Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 28 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 19 Nov, 2025 Reviews received at journal 18 Nov, 2025 Reviews received at journal 07 Nov, 2025 Reviewers agreed at journal 31 Oct, 2025 Reviewers agreed at journal 31 Oct, 2025 Reviewers agreed at journal 30 Oct, 2025 Reviewers agreed at journal 27 Oct, 2025 Reviewers invited by journal 27 Oct, 2025 Editor assigned by journal 24 Oct, 2025 Editor invited by journal 24 Oct, 2025 Submission checks completed at journal 16 Oct, 2025 First submitted to journal 16 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7808482","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":538076350,"identity":"77b58fef-449a-4667-bafc-92c7abaf5f18","order_by":0,"name":"Runmei Zhang","email":"","orcid":"","institution":"School of Electronic and Information Engineering, Anhui Jianzhu University","correspondingAuthor":false,"prefix":"","firstName":"Runmei","middleName":"","lastName":"Zhang","suffix":""},{"id":538076351,"identity":"814cebdf-d549-4c2e-aab1-34d79822ea35","order_by":1,"name":"Chenfei Pan","email":"","orcid":"","institution":"School of Electronic and Information Engineering, Anhui Jianzhu University","correspondingAuthor":false,"prefix":"","firstName":"Chenfei","middleName":"","lastName":"Pan","suffix":""},{"id":538076352,"identity":"99ff8749-9c1e-4be9-88c9-cbdd4bf43767","order_by":2,"name":"Zihua Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIie3QsUoDQRCA4TkWTouBtNskeYWRhVh4pPFF5ji4awyktDxIK7n2RB/i8gYbprCJ1ldYCEIqiwVbQVdtrDZnJ7h/MwzMB8sCxGJ/MgTrKJuABv5ck3oASbbtsjS/IkqOneS1/l4Pk1FzbwVIquamfn5FyMadVfunENH2gWVJ1aJ9tGwQStPZ9JRChJIrkpbOFnXPXCBI3llMdZAoJEFS1dQTQXgfQNIvcs7Uc77yH36Y6B3StqXyZNNzkdxSYa4lnQXJqNkZ596y6aS/KNzL5Xy8vlvtg+RHyP6dfqqB974jO/w2FovF/lUfBwRMopxlpo4AAAAASUVORK5CYII=","orcid":"","institution":"School of Electronic and Information Engineering, Anhui Jianzhu University","correspondingAuthor":true,"prefix":"","firstName":"Zihua","middleName":"","lastName":"Chen","suffix":""},{"id":538076353,"identity":"29396e2b-94d9-4fc8-b652-0094c40f93a4","order_by":3,"name":"Jingwei Fan","email":"","orcid":"","institution":"School of Mechanical and Electrical Engineering, Anhui Jianzhu University","correspondingAuthor":false,"prefix":"","firstName":"Jingwei","middleName":"","lastName":"Fan","suffix":""},{"id":538076354,"identity":"7ae74cd2-c106-4e34-83dc-92e795b496ba","order_by":4,"name":"Zhong Chen","email":"","orcid":"","institution":"School of Mechanical and Electrical Engineering, Anhui Jianzhu University","correspondingAuthor":false,"prefix":"","firstName":"Zhong","middleName":"","lastName":"Chen","suffix":""},{"id":538076355,"identity":"67a944be-2d7e-4e6c-96fd-a22bfc71c761","order_by":5,"name":"Bin Yuan","email":"","orcid":"","institution":"School of Mechanical and Electrical Engineering, Anhui Jianzhu University","correspondingAuthor":false,"prefix":"","firstName":"Bin","middleName":"","lastName":"Yuan","suffix":""}],"badges":[],"createdAt":"2025-10-08 13:38:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7808482/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7808482/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-41568-2","type":"published","date":"2026-02-28T15:59:14+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":95180331,"identity":"aac1d0df-bef8-4e97-8ae8-d78bf5dc0bcb","added_by":"auto","created_at":"2025-11-05 08:21:11","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7113,"visible":true,"origin":"","legend":"","description":"","filename":"fd72b0ef912e4b9899be974f9202c7be.json","url":"https://assets-eu.researchsquare.com/files/rs-7808482/v1/f56b670735d2886cfb792776.json"},{"id":103765807,"identity":"e9ddbad8-1cef-4d30-af41-fcea7d9d56f8","added_by":"auto","created_at":"2026-03-02 16:09:02","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2521236,"visible":true,"origin":"","legend":"","description":"","filename":"TemplateforsubmissionstoScientificReports2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7808482/v1_covered_88166bed-42e6-463d-8cd3-feb5d6fdcda8.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"An Improved Lightweight YOLOv11 Algorithm for Weld Surface Defect Detection","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Deep learning, Defect detection, YOLOv11, PConv","lastPublishedDoi":"10.21203/rs.3.rs-7808482/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7808482/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIndustrial welding often exhibits some essential problems, such as unclear defect characteristics and complex backgroundinformation. However, the existing defect detection models have relatively high costs and may be weak in weld defect detection.To address the problem, this paper proposes an improved lightweight YOLOv11 model for welding surface defect detection,called YOLO-Air. First, the model integrates the feature extraction module with the convolutional module to enhance featureextraction capabilities and optimize computational resource utilization. Second, the GSConv and VOV-GSCSP modules areembedded in the neck network to reduce feature redundancy in both spatial and channel dimensions, and then lower thecomputational load. Third, a lightweight detection head is designed as part of the detection network to further reduce modelcomplexity. Finally, we compare our proposed YOLO-Air model to the baseline model via the Welding Defect Test-V2 andNEU-DET dataset. Experimental results show that the proposed model achieves superior weld surface detection effects,especially the metric mAP50 is +1.3%, the parameter number is reduced by 17.3%, and the computational load is reduced by 31.7%\u003c/p\u003e","manuscriptTitle":"An Improved Lightweight YOLOv11 Algorithm for Weld Surface Defect Detection","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-05 08:21:07","doi":"10.21203/rs.3.rs-7808482/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-19T15:57:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-18T11:19:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-07T08:49:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"73581699948344590327017621677927571277","date":"2025-10-31T12:58:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"1811805134475960004399499589864978477","date":"2025-10-31T10:04:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"52987352821515084530961324009177036143","date":"2025-10-30T12:20:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"73686394458769416587100594402066655002","date":"2025-10-27T11:47:06+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-27T11:39:16+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-24T13:32:57+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-10-24T04:21:48+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-16T09:39:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-10-16T09:38:29+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"51084646-8149-4bf3-b183-baedf1a1d517","owner":[],"postedDate":"November 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":57228662,"name":"Physical sciences/Engineering"},{"id":57228663,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-03-02T16:06:03+00:00","versionOfRecord":{"articleIdentity":"rs-7808482","link":"https://doi.org/10.1038/s41598-026-41568-2","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-02-28 15:59:14","publishedOnDateReadable":"February 28th, 2026"},"versionCreatedAt":"2025-11-05 08:21:07","video":"","vorDoi":"10.1038/s41598-026-41568-2","vorDoiUrl":"https://doi.org/10.1038/s41598-026-41568-2","workflowStages":[]},"version":"v1","identity":"rs-7808482","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7808482","identity":"rs-7808482","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}