MSBNet: Handwritten Bangla Character Recognition Using Lightweight Multi-scale CNN Architecture

MSBNet: Handwritten Bangla Character Recognition Using Lightweight Multi-scale CNN Architecture | 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 MSBNet: Handwritten Bangla Character Recognition Using Lightweight Multi-scale CNN Architecture Rejoy Chakraborty, Chayan Halder, Kaushik Roy, Shivam Gupta, Shashi Shekhar Jha This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6825617/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract With the rise in machine learning deployments in many real-world applications, the demand for handwritten character recognition is increasing rapidly. Most of the applications are currently restricted to the English language, but to broaden their impact, it is also essential to accommodate non-English speakers. This has led to an urgent need for character recognition systems for regional languages. The past few years have witnessed work in regional languages worldwide, ranging from Chinese and French to Indic languages. Within the field of Indic languages, the work on Bangla character recognition is in very nascent stages. The primary challenge is the enriched character set of more than 80 classes (including compound characters) and the variation in handwriting styles across different subjects. Most existing works are either limited to a subset of classes or have significant computational or storage overheads. To address this, the present paper proposes a novel lightweight multi-scale convolutional neural network (CNN), called MSBNet, that generalizes well for a variety of real-world handwritten Bangla character datasets. Furthermore, MSBNet can be smoothly integrated into edge devices due to its significant reduction in the number of parameters (correspondingly storage and computational requirements) compared to state-of-the-art (SOTA) architectures. The paper validates the efficacy of the proposed model experimentally and conducts an ablation study on the model 1 Handwritten Bangla Character Classification Multi-scale Architecture CNN Deep Learning Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted 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-6825617","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":466802546,"identity":"a8cbb9cf-59a4-42b1-8147-9784f8cbc088","order_by":0,"name":"Rejoy Chakraborty","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/UlEQVRIie3RsUoDMRjA8Xx8kFsCWXOc+AyfdKgHhb5KXBy14NJJTg7icurqkxS6FTJ0uQco3HJ9A8UlYAZzWqwiqY4O+S8JfPnlCMdYKvUPk4i2D6vgCDe9ps+BipL81pwP545khjVpot8JtS0N00n+kJmwoejJfZtws/MLQRbMvJ/5y3GxgifHTi9iAh613t6Z7p0MvFzea8wbpq5iBJVejaDaE6KWsSK85ayKEB5GBfMfZLYj+HqICGHDnbwTeQ2G7Qg/+BWVGX4yvEUi1ErTiMoGTNlQnEytfOmd76ZcrrfPzh/TWKDduPl1lPyMGFR/+z9fSCqVSqW+9wYlKFFkfN0STAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-9584-1838","institution":"Indian Institute of Technology Ropar","correspondingAuthor":true,"prefix":"","firstName":"Rejoy","middleName":"","lastName":"Chakraborty","suffix":""},{"id":466802547,"identity":"b1879cc1-6abc-4c27-9688-12b353fbf511","order_by":1,"name":"Chayan Halder","email":"","orcid":"https://orcid.org/0000-0002-6113-7284","institution":"Ramakrishna Mission Vivekananda Centenary College","correspondingAuthor":false,"prefix":"","firstName":"Chayan","middleName":"","lastName":"Halder","suffix":""},{"id":466802548,"identity":"253be57d-cc82-41d2-8232-3521442ced9b","order_by":2,"name":"Kaushik Roy","email":"","orcid":"https://orcid.org/0000-0002-3360-7576","institution":"West Bengal State University","correspondingAuthor":false,"prefix":"","firstName":"Kaushik","middleName":"","lastName":"Roy","suffix":""},{"id":466802549,"identity":"23aec830-78da-4f06-adcc-5756e740095d","order_by":3,"name":"Shivam Gupta","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0003-0633-3917","institution":"Indian Institute of Technology Ropar","correspondingAuthor":true,"prefix":"","firstName":"Shivam","middleName":"","lastName":"Gupta","suffix":""},{"id":466802550,"identity":"8cbed617-9842-418c-8427-54c1e31a4f10","order_by":4,"name":"Shashi Shekhar Jha","email":"","orcid":"https://orcid.org/0000-0002-1375-2266","institution":"Indian Institute of Technology Ropar","correspondingAuthor":false,"prefix":"","firstName":"Shashi","middleName":"Shekhar","lastName":"Jha","suffix":""}],"badges":[],"createdAt":"2025-06-05 06:06:50","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6825617/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6825617/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":84039787,"identity":"997289c3-84f0-496b-8a0e-0cab95e4c90b","added_by":"auto","created_at":"2025-06-06 05:33:59","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":498925,"visible":true,"origin":"","legend":"","description":"","filename":"BanglaCharacterMultiScaleCODSCOMAD1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6825617/v1_covered_84dd3e7c-9f9a-4dea-a344-59b64747297e.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eMSBNet: Handwritten Bangla Character Recognition Using Lightweight Multi-scale CNN Architecture\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Indian Institute of Technology Ropar","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Handwritten Bangla Character, Classification, Multi-scale Architecture, CNN, Deep Learning","lastPublishedDoi":"10.21203/rs.3.rs-6825617/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6825617/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWith the rise in machine learning deployments in many real-world applications, the demand for handwritten character recognition is increasing rapidly. Most of the applications are currently restricted to the English language, but to broaden their impact, it is also essential to accommodate non-English speakers. This has led to an urgent need for character recognition systems for regional languages. The past few years have witnessed work in regional languages worldwide, ranging from Chinese and French to Indic languages. Within the field of Indic languages, the work on Bangla character recognition is in very nascent stages. The primary challenge is the enriched character set of more than 80 classes (including compound characters) and the variation in handwriting styles across different subjects. Most existing works are either limited to a subset of classes or have significant computational or storage overheads. To address this, the present paper proposes a novel lightweight multi-scale convolutional neural network (CNN), called MSBNet, that generalizes well for a variety of real-world handwritten Bangla character datasets. Furthermore, MSBNet can be smoothly integrated into edge devices due to its significant reduction in the number of parameters (correspondingly storage and computational requirements) compared to state-of-the-art (SOTA) architectures. The paper validates the efficacy of the proposed model experimentally and conducts an ablation study on the model\u003csup\u003e1\u003c/sup\u003e\u003c/p\u003e","manuscriptTitle":"MSBNet: Handwritten Bangla Character Recognition Using Lightweight Multi-scale CNN Architecture","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-06 05:09:52","doi":"10.21203/rs.3.rs-6825617/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"31c738ff-38d4-45ec-bb78-ee4a0d44edb4","owner":[],"postedDate":"June 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-06-06T05:09:52+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-06 05:09:52","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6825617","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6825617","identity":"rs-6825617","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}