A Hybrid Approach to Bangla Handwritten OCR: Combining YOLO and an Advanced CNN

preprint OA: closed CC-BY-4.0
📄 Open PDF Full text JSON View at publisher

Abstract

Abstract Optical Character Recognition (OCR) plays an important role in automatic data entry from handwritten forms into digital systems. However, there is a gap in the research on the development of OCR techniques for handwritten texts in complex languages such as Bangla. Complexities in Bangla writing arise from the presence of modifiers, compound characters, and diacritic marks. Our research focuses on developing a scalable and effective method for handwritten documents in Bangla that addresses the complexities of word images. The pipeline uses the YOLO (You Only Look Once) model to isolate characters, including base alphabets, consonant conjuncts and characters with modifiers(matras), from word images and a Deep Learning model for character recognition. The hybrid model proved to be highly accurate in identifying characters from more complex sets and different penmanship styles. In addition, the resilience of the system was enhanced using a Word2Vec-anchored spelling correction technology.
Full text 12,622 characters · extracted from preprint-html · click to expand
A Hybrid Approach to Bangla Handwritten OCR: Combining YOLO and an Advanced CNN | 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 A Hybrid Approach to Bangla Handwritten OCR: Combining YOLO and an Advanced CNN Aye Maung, Sumaiya Salekin, Mohammad A. Haque This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5149357/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 21 Jun, 2025 Read the published version in Discover Artificial Intelligence → Version 1 posted 12 You are reading this latest preprint version Abstract Optical Character Recognition (OCR) plays an important role in automatic data entry from handwritten forms into digital systems. However, there is a gap in the research on the development of OCR techniques for handwritten texts in complex languages such as Bangla. Complexities in Bangla writing arise from the presence of modifiers, compound characters, and diacritic marks. Our research focuses on developing a scalable and effective method for handwritten documents in Bangla that addresses the complexities of word images. The pipeline uses the YOLO (You Only Look Once) model to isolate characters, including base alphabets, consonant conjuncts and characters with modifiers(matras), from word images and a Deep Learning model for character recognition. The hybrid model proved to be highly accurate in identifying characters from more complex sets and different penmanship styles. In addition, the resilience of the system was enhanced using a Word2Vec-anchored spelling correction technology. Bangla OCR Character Detection Character Recognition Word Recognition YOLO EfficientNet Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 21 Jun, 2025 Read the published version in Discover Artificial Intelligence → Version 1 posted Editorial decision: Revision requested 07 Nov, 2024 Reviews received at journal 30 Oct, 2024 Reviews received at journal 25 Oct, 2024 Reviewers agreed at journal 23 Oct, 2024 Reviews received at journal 22 Oct, 2024 Reviewers agreed at journal 21 Oct, 2024 Reviewers agreed at journal 21 Oct, 2024 Reviewers agreed at journal 19 Oct, 2024 Reviewers invited by journal 15 Oct, 2024 Editor assigned by journal 08 Oct, 2024 Submission checks completed at journal 01 Oct, 2024 First submitted to journal 25 Sep, 2024 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-5149357","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":375334698,"identity":"b1cbb6ad-a7f0-411e-bdf5-2c9f37b5c6c1","order_by":0,"name":"Aye Maung","email":"","orcid":"","institution":"Bangladesh University of Engineering and Technology","correspondingAuthor":false,"prefix":"","firstName":"Aye","middleName":"","lastName":"Maung","suffix":""},{"id":375334699,"identity":"e9a4453b-86b8-49f2-ba43-24e92a10b571","order_by":1,"name":"Sumaiya Salekin","email":"","orcid":"","institution":"Bangladesh University of Engineering and Technology","correspondingAuthor":false,"prefix":"","firstName":"Sumaiya","middleName":"","lastName":"Salekin","suffix":""},{"id":375334700,"identity":"6ce21105-e00c-4495-82f5-a008261bfed6","order_by":2,"name":"Mohammad A. Haque","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEElEQVRIiWNgGAWjYFACxoYPDAUHwKwHINIAJmHAkIBLS+MMBgOwFmYDqBbGBvxaGBhhWtgkiNLCP/twYzOPwR05vhvJxyo+VNgwmEskH3/AUGPHYM6OXYvEuUSQlmfGkjfS0m7OOJPGYDkjLbGB4Vgyg2XPA+zOOsPY/pjH4HDihhs5Zrd52w4zGNzIMWxgYDsAZGC3Rf4MI8gWkJb8b8V///0Hqsz/2MDwD7cWA4SWHDZmxgaQyhzGBsY23FoMgVoa5xgcNpY8A/RPz7FkHsueZ4YzEvtADOx+kTvD/rDhTcVhOb7jyQ8//KixkzNnT37w4cM3EAO7LQhwAELxgMkEGIMYLaNgFIyCUTAKMAAA995oHB6QK38AAAAASUVORK5CYII=","orcid":"","institution":"Bangladesh University of Engineering and Technology","correspondingAuthor":true,"prefix":"","firstName":"Mohammad","middleName":"A.","lastName":"Haque","suffix":""}],"badges":[],"createdAt":"2024-09-25 06:23:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5149357/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5149357/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s44163-025-00251-7","type":"published","date":"2025-06-21T15:57:29+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":85231333,"identity":"e14b439c-b9af-4103-9c42-25a37e84ef38","added_by":"auto","created_at":"2025-06-23 16:06:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1139497,"visible":true,"origin":"","legend":"","description":"","filename":"AHybridApproachtoBanglaHandwrittenOCRCombiningYOLOandanAdvancedCNN.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5149357/v1_covered_b11eb196-d968-4736-98e3-4f9cbd8092c9.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Hybrid Approach to Bangla Handwritten OCR: Combining YOLO and an Advanced CNN","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":"discover-artificial-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"diai","sideBox":"Learn more about [Discover Artificial Intelligence](https://www.springer.com/44163)","snPcode":"","submissionUrl":"","title":"Discover Artificial Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Bangla OCR, Character Detection, Character Recognition, Word Recognition, YOLO, EfficientNet","lastPublishedDoi":"10.21203/rs.3.rs-5149357/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5149357/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Optical Character Recognition (OCR) plays an important role in automatic data entry from handwritten forms into digital systems. However, there is a gap in the research on the development of OCR techniques for handwritten texts in complex languages such as Bangla. Complexities in Bangla writing arise from the presence of modifiers, compound characters, and diacritic marks. Our research focuses on developing a scalable and effective method for handwritten documents in Bangla that addresses the complexities of word images. The pipeline uses the YOLO (You Only Look Once) model to isolate characters, including base alphabets, consonant conjuncts and characters with modifiers(matras), from word images and a Deep Learning model for character recognition. The hybrid model proved to be highly accurate in identifying characters from more complex sets and different penmanship styles. In addition, the resilience of the system was enhanced using a Word2Vec-anchored spelling correction technology.","manuscriptTitle":"A Hybrid Approach to Bangla Handwritten OCR: Combining YOLO and an Advanced CNN","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-11-19 15:05:34","doi":"10.21203/rs.3.rs-5149357/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-11-07T13:17:16+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-30T04:12:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-25T10:17:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"137840263184590882712949590400366082547","date":"2024-10-23T16:02:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-22T06:34:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"188425631739131301624743150525376778150","date":"2024-10-21T15:58:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"5273240564410852438135939195618069704","date":"2024-10-21T15:21:53+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"266726937649100093567755601493588770445","date":"2024-10-19T12:01:06+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-10-16T02:45:56+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-10-08T07:45:20+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-10-01T12:44:58+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Artificial Intelligence","date":"2024-09-25T06:18:26+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-artificial-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"diai","sideBox":"Learn more about [Discover Artificial Intelligence](https://www.springer.com/44163)","snPcode":"","submissionUrl":"","title":"Discover Artificial Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1a514b99-2a37-4499-b091-c095b1709828","owner":[],"postedDate":"November 19th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-06-23T16:00:54+00:00","versionOfRecord":{"articleIdentity":"rs-5149357","link":"https://doi.org/10.1007/s44163-025-00251-7","journal":{"identity":"discover-artificial-intelligence","isVorOnly":false,"title":"Discover Artificial Intelligence"},"publishedOn":"2025-06-21 15:57:29","publishedOnDateReadable":"June 21st, 2025"},"versionCreatedAt":"2024-11-19 15:05:34","video":"","vorDoi":"10.1007/s44163-025-00251-7","vorDoiUrl":"https://doi.org/10.1007/s44163-025-00251-7","workflowStages":[]},"version":"v1","identity":"rs-5149357","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5149357","identity":"rs-5149357","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2024) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-05-26T02:00:01.498150+00:00
License: CC-BY-4.0