APAU-Net: Adaptive Prior-Aware U-Net Text-Line Segmentation for Historical Documents

APAU-Net: Adaptive Prior-Aware U-Net Text-Line Segmentation for Historical Documents | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || '';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a Preprint Cite Share Download PDF Research Article APAU-Net: Adaptive Prior-Aware U-Net Text-Line Segmentation for Historical Documents Mohamed Amine Beghoura, Abdelouahab Attia, Abderraouf Bouziane, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8202596/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Text-line segmentation in historical manuscripts remains challenging due to degradation, overlapping strokes, and extreme data scarcity. We propose APAU-Net, a two-stage cascade architecture that explicitly optimizes Line Intersection-over-Union (Line IU) via learned anisotropic Gaussian priors. Stage 1 predicts topology-aware ellipsoidal priors from low resolution grayscale images using connected-component analysis and moment-based ellipse fitting. Stage 2 refines these priors at full resolution through a residual U-Net with adaptive per-pixel weighting and gated fusion. We evaluated APAU-Net on the challenging U-DIADS-TL (84 images, only 3 training pages per manuscript) and DIVA-HisDB benchmarks. It achieves an average Line Intersection-over-Union (Line IU) of 94.3% (+9.8 pp over plain U-Net) and outperforms recent few-shot baselines by up to +24 pp on the most degraded Syriac subset. Ablation confirms the anisotropic prior contributes ~9-12 pp to Line IU under severe data scarcity. The source code for the proposed method is openly available on GitHub at \href{ https://github.com/aminebeg/APAUNet}{https://github.com/aminebeg/APAUNet} Document image analysis Historical manuscripts Text line segmentation Deep learning Prior-guided segmentation Oriented Gaussian prior Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 04 May, 2026 Reviews received at journal 16 Apr, 2026 Reviews received at journal 21 Feb, 2026 Reviewers agreed at journal 13 Feb, 2026 Reviewers agreed at journal 08 Jan, 2026 Reviewers invited by journal 10 Dec, 2025 Editor assigned by journal 26 Nov, 2025 Submission checks completed at journal 26 Nov, 2025 First submitted to journal 25 Nov, 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-8202596","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":558478949,"identity":"32e4815c-c75d-49b2-934b-d556d3e5511b","order_by":0,"name":"Mohamed Amine Beghoura","email":"data:image/png;base64,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","orcid":"","institution":"University Mohamed El Bachir El Ibrahimi of Bordj Bou Arreridj","correspondingAuthor":true,"prefix":"","firstName":"Mohamed","middleName":"Amine","lastName":"Beghoura","suffix":""},{"id":558478950,"identity":"145c2c81-9701-48b3-a325-0f54275a96a2","order_by":1,"name":"Abdelouahab Attia","email":"","orcid":"","institution":"University Mohamed El Bachir El Ibrahimi of Bordj Bou Arreridj","correspondingAuthor":false,"prefix":"","firstName":"Abdelouahab","middleName":"","lastName":"Attia","suffix":""},{"id":558478951,"identity":"25ebcc40-a22a-49dc-900c-6cbeb6579059","order_by":2,"name":"Abderraouf Bouziane","email":"","orcid":"","institution":"University Mohamed El Bachir El Ibrahimi of Bordj Bou Arreridj","correspondingAuthor":false,"prefix":"","firstName":"Abderraouf","middleName":"","lastName":"Bouziane","suffix":""},{"id":558478952,"identity":"270b0b64-0da2-4f53-8331-b77a570987aa","order_by":3,"name":"M. Hassaballah","email":"","orcid":"","institution":"Prince Sattam Bin Abdulaziz University","correspondingAuthor":false,"prefix":"","firstName":"M.","middleName":"","lastName":"Hassaballah","suffix":""}],"badges":[],"createdAt":"2025-11-25 11:23:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8202596/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8202596/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98265763,"identity":"5847a156-b559-4088-a7aa-f1540153cfcb","added_by":"auto","created_at":"2025-12-15 22:39:20","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5297,"visible":true,"origin":"","legend":"","description":"","filename":"cff9f319420a45c6b3c789bc86967b9c.json","url":"https://assets-eu.researchsquare.com/files/rs-8202596/v1/591e57271a3f57184d0f0d82.json"},{"id":98435525,"identity":"bad9a982-68c9-4eef-9b67-48622b3d5eeb","added_by":"auto","created_at":"2025-12-17 16:54:00","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":48971401,"visible":true,"origin":"","legend":"","description":"","filename":"APAU.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8202596/v1_covered_7c95353b-fc37-49d0-a723-420eaed1bace.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"APAU-Net: Adaptive Prior-Aware U-Net Text-Line Segmentation for Historical Documents","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":"international-journal-on-document-analysis-and-recognition-ijdar","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ijda","sideBox":"Learn more about [International Journal on Document Analysis and Recognition (IJDAR)](http://link.springer.com/journal/10032)","snPcode":"10032","submissionUrl":"https://submission.nature.com/new-submission/10032/3","title":"International Journal on Document Analysis and Recognition (IJDAR)","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Document image analysis, Historical manuscripts, Text line segmentation, Deep learning, Prior-guided segmentation, Oriented Gaussian prior","lastPublishedDoi":"10.21203/rs.3.rs-8202596/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8202596/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Text-line segmentation in historical manuscripts remains challenging due to degradation, overlapping strokes, and extreme data scarcity. We propose APAU-Net, a two-stage cascade architecture that explicitly optimizes Line Intersection-over-Union (Line IU) via learned anisotropic Gaussian priors. Stage 1 predicts topology-aware ellipsoidal priors from low resolution grayscale images using connected-component analysis and moment-based ellipse fitting. Stage 2 refines these priors at full resolution through a residual U-Net with adaptive per-pixel weighting and gated fusion. We evaluated APAU-Net on the challenging U-DIADS-TL (84 images, only 3 training pages per manuscript) and DIVA-HisDB benchmarks. It achieves an average Line Intersection-over-Union (Line IU) of 94.3% (+9.8 pp over plain U-Net) and outperforms recent few-shot baselines by up to +24 pp on the most degraded Syriac subset. Ablation confirms the anisotropic prior contributes ~9-12 pp to Line IU under severe data scarcity. The source code for the proposed method is openly available on GitHub at \\href{https://github.com/aminebeg/APAUNet}{https://github.com/aminebeg/APAUNet}","manuscriptTitle":"APAU-Net: Adaptive Prior-Aware U-Net Text-Line Segmentation for Historical Documents","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-15 22:39:16","doi":"10.21203/rs.3.rs-8202596/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-04T13:47:57+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-16T08:10:46+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-21T20:14:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"126326470832775625777102654764240674022","date":"2026-02-13T10:50:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"51876767962658976729323934698741035150","date":"2026-01-08T09:01:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-10T13:39:48+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-27T02:22:10+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-26T09:57:27+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal on Document Analysis and Recognition (IJDAR)","date":"2025-11-25T11:11:49+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"international-journal-on-document-analysis-and-recognition-ijdar","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ijda","sideBox":"Learn more about [International Journal on Document Analysis and Recognition (IJDAR)](http://link.springer.com/journal/10032)","snPcode":"10032","submissionUrl":"https://submission.nature.com/new-submission/10032/3","title":"International Journal on Document Analysis and Recognition (IJDAR)","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"13e8cfd1-9282-44a4-af20-cc85211d2e1b","owner":[],"postedDate":"December 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-26T15:10:27+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-15 22:39:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8202596","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8202596","identity":"rs-8202596","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}