Self-Supervised Plankton Classification via DINO and Gradient-Based Loss Re-weighting

Self-Supervised Plankton Classification via DINO and Gradient-Based Loss Re-weighting | 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 Self-Supervised Plankton Classification via DINO and Gradient-Based Loss Re-weighting Abdelfatah Ahmad, Zaid Habash, Abdulrahman Ahmad, Muaz Alradi, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9109751/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 Manual identification of plankton species is labor intensive and impractical for large-scale ecological analysis. Deep learning offers a promising solution but suffers when trained on small and imbalanced datasets, common in microscopic image domains. In this paper, we present a comparative evaluation of several convolutional neural networks (CNNs), including VGG16 and DenseNet201, and propose a novel approach integrating a DINO vision transformer for feature extraction. Additionally, we introduce a gradient-based loss weighting mechanism to estimate the effective number of samples per class, improving training stability under imbalance. Experimental results on a 19-class plankton dataset show that our DINO-based model achieves a 97% test accuracy, significantly outperforming all CNN baselines. These findings highlight the effectiveness of self- supervised transformers and tailored loss strategies for robust plankton classification. Aquaculture and Mariculture Plankton classification DINO Vision Trans- former Imbalanced data Gradient-based Re-weighting 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-9109751","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":605396652,"identity":"33be318f-3652-4d1d-9f6f-f31f00ac076f","order_by":0,"name":"Abdelfatah Ahmad","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2UlEQVRIiWNgGAWjYBACNmbGhsM/Kmzk+NnbiNTCx9588DHDmTRjyZ5jRGqR4zmWbMzYdihxw400Yh0mkWMmXdh2wNjg5rM0CYYaOwb+9gYitMw4d0dO8nbaMQmGY8kMEmcOENYiwVP2zJjvdnqbBAMbUPmNBGK0sB1ObLh5HKjl3wEG+fsPCGgBeZ+n7XDihBtsxyQY2w4wGNzAr4OBDRjID2eAAzkt2SKxL5nH8AwBh8k3MzYc+ACOymOGNz58s5OTO36AgDUoAGg+DynqR8EoGAWjYBTgAAD1h0YTM5wdJwAAAABJRU5ErkJggg==","orcid":"","institution":"Khalifa University","correspondingAuthor":true,"prefix":"","firstName":"Abdelfatah","middleName":"","lastName":"Ahmad","suffix":""},{"id":605396653,"identity":"befccef1-6590-4bbd-a144-ff1da869022b","order_by":1,"name":"Zaid Habash","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Zaid","middleName":"","lastName":"Habash","suffix":""},{"id":605396654,"identity":"242a48d0-a90f-449d-942f-4781f2ca1b02","order_by":2,"name":"Abdulrahman Ahmad","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Abdulrahman","middleName":"","lastName":"Ahmad","suffix":""},{"id":605396655,"identity":"1c860042-84ac-45a2-b639-044cd3ee048a","order_by":3,"name":"Muaz Alradi","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Muaz","middleName":"","lastName":"Alradi","suffix":""},{"id":605396656,"identity":"e6f1da9d-1874-445f-bb52-77f7e5973d96","order_by":4,"name":"Abderrahmene Boudiaf","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Abderrahmene","middleName":"","lastName":"Boudiaf","suffix":""},{"id":605396657,"identity":"160d3337-8fe5-43fb-8379-629a7469b31e","order_by":5,"name":"Naoufel Werghi","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Naoufel","middleName":"","lastName":"Werghi","suffix":""},{"id":605396658,"identity":"856fd069-9778-4a2b-bdef-6d7283d01a0c","order_by":6,"name":"Muhammad Owais","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Muhammad","middleName":"","lastName":"Owais","suffix":""},{"id":605396659,"identity":"7ccbd0f4-c0b7-4614-a700-eaba505637d9","order_by":7,"name":"Irfan Hussain","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Irfan","middleName":"","lastName":"Hussain","suffix":""},{"id":605396660,"identity":"b1dcda79-f8e9-43d8-a0c0-3841ce99d95c","order_by":8,"name":"Taimur Hassan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYFACxgYQmQDhVEC5+AAPqpYzRGmBAIgWxjYitNizH27dzLuHIY+///jDz4Xz7GQ3HGB++IGhxga3LTyJbbd5njEUS9zIMZaeuS3ZeMMBNmMJhmNpeBwG0nKAIbHhBg+DNO825sQNBxjMGBjYDuPWwv8QomX++eOPf/POqQdqYf/GwPDvP24tElBbNhxIMJPmbTgMZPCYAcPhAG4tNx623ZxzQCJx440cM2ueY8eNZx7mKZZI7EvGqYW9P/3ZjTcHbBLnAR12m6emWrbvePvGDx++2eHUAgUSSGxmBnjMjoJRMApGwSggEwAAAflXe8WX+QMAAAAASUVORK5CYII=","orcid":"","institution":"Abu Dhabi University","correspondingAuthor":true,"prefix":"","firstName":"Taimur","middleName":"","lastName":"Hassan","suffix":""}],"badges":[],"createdAt":"2026-03-13 03:29:13","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-9109751/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9109751/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104782259,"identity":"0919b8f7-d472-42f7-852d-80718b0768df","added_by":"auto","created_at":"2026-03-17 07:57:04","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":666591,"visible":true,"origin":"","legend":"","description":"","filename":"Paper2AbdulFateh.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9109751/v1_covered_164d9704-a1cc-477b-a2c2-c724f3af6179.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eSelf-Supervised Plankton Classification via DINO and Gradient-Based Loss Re-weighting\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Khalifa University of Science and Technology","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":"Plankton classification, DINO, Vision Trans- former, Imbalanced data, Gradient-based Re-weighting","lastPublishedDoi":"10.21203/rs.3.rs-9109751/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9109751/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eManual identification of plankton species is labor intensive and impractical for large-scale ecological analysis. Deep learning offers a promising solution but suffers when trained on small and imbalanced datasets, common in microscopic image domains. In this paper, we present a comparative evaluation of several convolutional neural networks (CNNs), including VGG16 and DenseNet201, and propose a novel approach integrating a DINO vision transformer for feature extraction. Additionally, we introduce a gradient-based loss weighting mechanism to estimate the effective number of samples per class, improving training stability under imbalance. Experimental results on a 19-class plankton dataset show that our DINO-based model achieves a 97% test accuracy, significantly outperforming all CNN baselines. These findings highlight the effectiveness of self- supervised transformers and tailored loss strategies for robust plankton classification.\u003c/p\u003e","manuscriptTitle":"Self-Supervised Plankton Classification via DINO and Gradient-Based Loss Re-weighting","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-16 01:12:34","doi":"10.21203/rs.3.rs-9109751/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":"3df705bd-ed6d-4cd5-8489-8fac2f0f1e31","owner":[],"postedDate":"March 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":64429426,"name":"Aquaculture and Mariculture"}],"tags":[],"updatedAt":"2026-03-16T01:12:34+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-16 01:12:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9109751","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9109751","identity":"rs-9109751","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}