Research on L-PBF Defect Detection Based on VAE-GAN Data Augmentation with Multi-Head Attention Mechanism

Research on L-PBF Defect Detection Based on VAE-GAN Data Augmentation with Multi-Head Attention Mechanism | 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 Research Article Research on L-PBF Defect Detection Based on VAE-GAN Data Augmentation with Multi-Head Attention Mechanism Yanhui Ma, Lusheng Li, Chengke Wang, Shuo Li, Bin Fu, Zhiqiong Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7036982/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 19 Feb, 2026 Read the published version in The International Journal of Advanced Manufacturing Technology → Version 1 posted 4 You are reading this latest preprint version Abstract Additive Manufacturing (AM), particularly Laser Powder Bed Fusion (L-PBF), faces critical challenges in defect detection due to the scarcity of high-quality training data and severe class imbalance, which significantly degrade the accuracy of deep learning models. To address these issues, this study proposes a novel data augmentation framework combining geometric transformations with an enhanced Variational Autoencoder-Generative Adversarial Network (VAE-GAN). Traditional augmentation techniques (rotation, scaling, flipping) are first applied to alleviate sample imbalance, followed by the improved VAE-GAN to synthesize high-fidelity defect images, thereby enriching dataset diversity. Experimental results on an L-PBF defect dataset demonstrate significant improvements in detection performance: This study conducted a comparative experiment evaluating the performance of YOLOv4, YOLOv7, YOLOv8, SSD, and Faster R-CNN on defect detection tasks before and after data augmentation. The results demonstrated significant mAP improvements across all models, with YOLOv4 achieving the most substantial enhancement (+20.03%, from 65.18% to 85.21%) despite its lower baseline performance. Faster R-CNN attained the highest post-augmentation mAP (87.75%), representing the best overall performance. YOLOv8 exhibited an optimal balance between real-time processing and accuracy (67.07%→84.66%, +17.59%), approaching Faster R-CNN's performance level. While SSD showed the smallest improvement (+12.34%), it maintained a relatively high baseline mAP (73.99%). These results validate the effectiveness of the proposed method in overcoming data scarcity and improving defect detection accuracy in AM. Additive Manufacturing Data Augmentation Small-Sample Learning Defect Detection Full Text Cite Share Download PDF Status: Published Journal Publication published 19 Feb, 2026 Read the published version in The International Journal of Advanced Manufacturing Technology → Version 1 posted Reviewers agreed at journal 09 Aug, 2025 Reviewers invited by journal 06 Aug, 2025 Editor assigned by journal 06 Aug, 2025 First submitted to journal 02 Aug, 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-7036982","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":496587527,"identity":"f9dc1c53-6d19-4d9e-9ba0-2bd66f26c7c3","order_by":0,"name":"Yanhui Ma","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yanhui","middleName":"","lastName":"Ma","suffix":""},{"id":496587528,"identity":"2110348d-19bc-4389-a1a6-784dda844866","order_by":1,"name":"Lusheng Li","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Lusheng","middleName":"","lastName":"Li","suffix":""},{"id":496587529,"identity":"099c3fbd-395c-4960-9d05-fdf8d5e20679","order_by":2,"name":"Chengke Wang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Chengke","middleName":"","lastName":"Wang","suffix":""},{"id":496587530,"identity":"5beed10e-b073-44f9-8f10-a9eef57cdda1","order_by":3,"name":"Shuo Li","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Shuo","middleName":"","lastName":"Li","suffix":""},{"id":496587531,"identity":"ac9b9605-e456-475f-8302-7c0cb4fd7bbe","order_by":4,"name":"Bin Fu","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Bin","middleName":"","lastName":"Fu","suffix":""},{"id":496587532,"identity":"b7fb8624-33e9-4772-ab09-9f00d280c763","order_by":5,"name":"Zhiqiong Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIiWNgGAWjYPCCGh779gYGAyCLsYFILcdkDHgOk6aF2cZAIhnMIqxFvj3H8HPBLzYec8n3B4p5GGxkNxxgfvYAnxbGnjfG0jP7ZHgsZyczGPMwpBlvOMBmboDXRRK5G6R5e9h4GG6DtRxO3HCAh00CnxY2idzNv3l7mHkYbh4GaflPWAuPRO42aZ4fzDwGN5hBWg4Q1iLB8/6bNW/DMR7JnmQDwzkGycYzD7OZ4dUi356WfJvnT409P/vBZwZvKuxk+443P8OrhYEhARhsbRB/GYAjkxm/eogWhj9gFvMDgopHwSgYBaNgRAIALC5B5ldG+YAAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0003-1430-0795","institution":"Tianjin University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Zhiqiong","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-07-03 10:27:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7036982/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7036982/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00170-026-17500-3","type":"published","date":"2026-02-19T15:59:45+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":103251724,"identity":"c3373657-a889-47ae-98fd-9a8d8e50595c","added_by":"auto","created_at":"2026-02-23 16:11:34","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":953020,"visible":true,"origin":"","legend":"","description":"","filename":"ResearchonLPBFDefectDetection.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7036982/v1_covered_16bd6a46-7d29-40d7-9bed-a08a8c3afe0f.pdf"}],"financialInterests":"","formattedTitle":"Research on L-PBF Defect Detection Based on VAE-GAN Data Augmentation with Multi-Head Attention Mechanism","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":true,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"the-international-journal-of-advanced-manufacturing-technology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jamt","sideBox":"Learn more about [The International Journal of Advanced Manufacturing Technology](https://www.springer.com/journal/170)","snPcode":"170","submissionUrl":"https://submission.nature.com/new-submission/170/3","title":"The International Journal of Advanced Manufacturing Technology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Additive Manufacturing, Data Augmentation, Small-Sample Learning, Defect Detection","lastPublishedDoi":"10.21203/rs.3.rs-7036982/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7036982/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Additive Manufacturing (AM), particularly Laser Powder Bed Fusion (L-PBF), faces critical challenges in defect detection due to the scarcity of high-quality training data and severe class imbalance, which significantly degrade the accuracy of deep learning models. To address these issues, this study proposes a novel data augmentation framework combining geometric transformations with an enhanced Variational Autoencoder-Generative Adversarial Network (VAE-GAN). Traditional augmentation techniques (rotation, scaling, flipping) are first applied to alleviate sample imbalance, followed by the improved VAE-GAN to synthesize high-fidelity defect images, thereby enriching dataset diversity. Experimental results on an L-PBF defect dataset demonstrate significant improvements in detection performance: This study conducted a comparative experiment evaluating the performance of YOLOv4, YOLOv7, YOLOv8, SSD, and Faster R-CNN on defect detection tasks before and after data augmentation. The results demonstrated significant mAP improvements across all models, with YOLOv4 achieving the most substantial enhancement (+20.03%, from 65.18% to 85.21%) despite its lower baseline performance. Faster R-CNN attained the highest post-augmentation mAP (87.75%), representing the best overall performance. YOLOv8 exhibited an optimal balance between real-time processing and accuracy (67.07%→84.66%, +17.59%), approaching Faster R-CNN's performance level. While SSD showed the smallest improvement (+12.34%), it maintained a relatively high baseline mAP (73.99%). These results validate the effectiveness of the proposed method in overcoming data scarcity and improving defect detection accuracy in AM.","manuscriptTitle":"Research on L-PBF Defect Detection Based on VAE-GAN Data Augmentation with Multi-Head Attention Mechanism","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-11 07:39:09","doi":"10.21203/rs.3.rs-7036982/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2025-08-09T17:28:53+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-06T12:13:13+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-06T09:06:27+00:00","index":"","fulltext":""},{"type":"submitted","content":"The International Journal of Advanced Manufacturing Technology","date":"2025-08-03T03:36:59+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"the-international-journal-of-advanced-manufacturing-technology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jamt","sideBox":"Learn more about [The International Journal of Advanced Manufacturing Technology](https://www.springer.com/journal/170)","snPcode":"170","submissionUrl":"https://submission.nature.com/new-submission/170/3","title":"The International Journal of Advanced Manufacturing Technology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"692f1401-691a-4722-9df2-58e6481ef158","owner":[],"postedDate":"August 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-02-23T16:08:08+00:00","versionOfRecord":{"articleIdentity":"rs-7036982","link":"https://doi.org/10.1007/s00170-026-17500-3","journal":{"identity":"the-international-journal-of-advanced-manufacturing-technology","isVorOnly":false,"title":"The International Journal of Advanced Manufacturing Technology"},"publishedOn":"2026-02-19 15:59:45","publishedOnDateReadable":"February 19th, 2026"},"versionCreatedAt":"2025-08-11 07:39:09","video":"","vorDoi":"10.1007/s00170-026-17500-3","vorDoiUrl":"https://doi.org/10.1007/s00170-026-17500-3","workflowStages":[]},"version":"v1","identity":"rs-7036982","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7036982","identity":"rs-7036982","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}