SF-YOLO11: A Real-Time Winter Jujube Detection Model Based on Lightweight Multi-Scale Fusion

SF-YOLO11: A Real-Time Winter Jujube Detection Model Based on Lightweight Multi-Scale Fusion | 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 SF-YOLO11: A Real-Time Winter Jujube Detection Model Based on Lightweight Multi-Scale Fusion Jichao Wang, Zhenqiao Hui, Mengyuan Li, Donglai Sun, Sicong Pang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8062910/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 To address the challenges of detecting winter jujubes in orchard environments—including dense small targets, variable illumination, complex backgrounds, and limited edge deployment capabilities—this paper proposes SF-YOLO11, a lightweight real-time detection model. First, a ternary channel importance metric is proposed that integrates gradient, variance, and task relevance to construct an adaptive pruning strategy. This strategy optimizes C3K2 into PrunedC3K2, achieving 31% model compression. Second, the LightSPPF module is designed through computational graph reconstruction. This module employs a three-stage architecture comprising depthwise separable convolution, adaptive pooling, and dynamic fusion, reducing computational cost by 40%. Finally, we transform multi-scale feature processing into sequence modeling and design the ISSFF module to capture scale dependencies through bidirectional LSTM and multi-head attention mechanisms, thereby improving small target detection accuracy. Experimental results demonstrate that SF-YOLO11 achieves an [email protected] of 92.89%, representing a 3.68% improvement over the baseline. The model contains 1.6M parameters, requires 4.3 GFLOPs of computation, and operates at 128 FPS, achieving optimizations of 38.5%, 33.8%, and 14%, respectively. After INT8 quantization, the model size is reduced to only 1.7 MB, and the FPS increases to 215. In cross-fruit transfer experiments, the model achieves zero-shot [email protected] values of 68.34% and 71.52% on grape and cherry tomato datasets, respectively. After fine-tuning, these values improve to 85.23% and 87.12%, respectively. Robustness tests validate the model's performance under extreme conditions, including strong illumination, low illumination, and occlusion. Winter jujube detection Lightweight deep learning Structured pruning Multi-scale feature fusion Small target detection Edge deployment YOLO11 Full Text Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.pdf Supplementary material 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-8062910","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":541887371,"identity":"69914d4d-24a6-478a-9810-b6e5f84e19c2","order_by":0,"name":"Jichao Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYLACCQOGBAYG5mPMEG4C0VrY0kjQAlHGY0acFoPjZw+/sCiwy+Nv7/n2uLDNhoGfPceA4ecOPFrO5KVZSBgkF0ucObvdeGZbGoNkzxsDxt4zuLWYHcgxM5AwOJDYcCN3mzTvtsMMBjdyDJgZ2/BoOf8GomX+jZxnQC3/GewJarmRY/wApGXDjRw2oJYDDAYSBLTY33hjBgzk5MSNZ46ZSc/8l8wjceZZwcFePFok+3OMP0v8sUucd7z5mXTBGTs5/vbkjQ9+4tECBGzSEkg8HhBxAK8GYEL5+IGAilEwCkbBKBjhAABWZlJt4AyPBQAAAABJRU5ErkJggg==","orcid":"","institution":"Hebei University of Water Resources and Electric Engineering","correspondingAuthor":true,"prefix":"","firstName":"Jichao","middleName":"","lastName":"Wang","suffix":""},{"id":541887372,"identity":"8a84efb3-784d-4ec3-9f36-8be0376a6b20","order_by":1,"name":"Zhenqiao Hui","email":"","orcid":"","institution":"Hebei University of Water Resources and Electric Engineering","correspondingAuthor":false,"prefix":"","firstName":"Zhenqiao","middleName":"","lastName":"Hui","suffix":""},{"id":541887373,"identity":"4500c80e-0316-4c9d-b3bd-c779fd6b32fb","order_by":2,"name":"Mengyuan Li","email":"","orcid":"","institution":"Hebei University of Water Resources and Electric Engineering","correspondingAuthor":false,"prefix":"","firstName":"Mengyuan","middleName":"","lastName":"Li","suffix":""},{"id":541887374,"identity":"8dcc4bee-7fe1-4273-a951-01250e28e7a4","order_by":3,"name":"Donglai Sun","email":"","orcid":"","institution":"Hebei University of Water Resources and Electric Engineering","correspondingAuthor":false,"prefix":"","firstName":"Donglai","middleName":"","lastName":"Sun","suffix":""},{"id":541887375,"identity":"164df19e-46c9-4cbb-91e4-808f3a2bd74d","order_by":4,"name":"Sicong Pang","email":"","orcid":"","institution":"Hebei University of Water Resources and Electric Engineering","correspondingAuthor":false,"prefix":"","firstName":"Sicong","middleName":"","lastName":"Pang","suffix":""}],"badges":[],"createdAt":"2025-11-08 09:23:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8062910/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8062910/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95635342,"identity":"b7933216-3d64-4895-9576-570148714550","added_by":"auto","created_at":"2025-11-11 12:21:39","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7527,"visible":true,"origin":"","legend":"","description":"","filename":"f3ab9495a1bf4107adeaa488dc2152b7.json","url":"https://assets-eu.researchsquare.com/files/rs-8062910/v1/a1a055beea31c95a13738906.json"},{"id":96248629,"identity":"c53c28c3-0e94-4ee5-b720-7b8e824bfc49","added_by":"auto","created_at":"2025-11-19 07:28:50","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":81508814,"visible":true,"origin":"","legend":"","description":"","filename":"SpringerNatureLaTeXTemplate177.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8062910/v1_covered_0248fb4b-4c1b-46d3-9ce2-d95724a0386f.pdf"},{"id":95635343,"identity":"6813661d-83cd-4c9b-9c3e-de1f93a4af94","added_by":"auto","created_at":"2025-11-11 12:21:42","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":82074794,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary material\u003c/p\u003e","description":"","filename":"SupplementaryMaterials.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8062910/v1/c52a5140697bf56d35e064a2.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"SF-YOLO11: A Real-Time Winter Jujube Detection Model Based on Lightweight Multi-Scale Fusion","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","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":"Winter jujube detection, Lightweight deep learning, Structured pruning, Multi-scale feature fusion, Small target detection, Edge deployment, YOLO11","lastPublishedDoi":"10.21203/rs.3.rs-8062910/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8062910/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"To address the challenges of detecting winter jujubes in orchard environments—including dense small targets, variable illumination, complex backgrounds, and limited edge deployment capabilities—this paper proposes SF-YOLO11, a lightweight real-time detection model. First, a ternary channel importance metric is proposed that integrates gradient, variance, and task relevance to construct an adaptive pruning strategy. This strategy optimizes C3K2 into PrunedC3K2, achieving 31% model compression. Second, the LightSPPF module is designed through computational graph reconstruction. This module employs a three-stage architecture comprising depthwise separable convolution, adaptive pooling, and dynamic fusion, reducing computational cost by 40%. Finally, we transform multi-scale feature processing into sequence modeling and design the ISSFF module to capture scale dependencies through bidirectional LSTM and multi-head attention mechanisms, thereby improving small target detection accuracy. Experimental results demonstrate that SF-YOLO11 achieves an [email protected] of 92.89%, representing a 3.68% improvement over the baseline. The model contains 1.6M parameters, requires 4.3 GFLOPs of computation, and operates at 128 FPS, achieving optimizations of 38.5%, 33.8%, and 14%, respectively. After INT8 quantization, the model size is reduced to only 1.7 MB, and the FPS increases to 215. In cross-fruit transfer experiments, the model achieves zero-shot [email protected] values of 68.34% and 71.52% on grape and cherry tomato datasets, respectively. After fine-tuning, these values improve to 85.23% and 87.12%, respectively. Robustness tests validate the model's performance under extreme conditions, including strong illumination, low illumination, and occlusion.","manuscriptTitle":"SF-YOLO11: A Real-Time Winter Jujube Detection Model Based on Lightweight Multi-Scale Fusion","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-11 12:21:34","doi":"10.21203/rs.3.rs-8062910/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":"c7fd740e-e86f-4335-9426-159274e85891","owner":[],"postedDate":"November 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-24T00:19:22+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-11 12:21:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8062910","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8062910","identity":"rs-8062910","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}