AD-YOLO: A unified method for traffic dense and small object detection in UAV images

AD-YOLO: A unified method for traffic dense and small object detection in UAV images | 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 AD-YOLO: A unified method for traffic dense and small object detection in UAV images Yu Deng, Yucong Hu, Yun Ye, Tiantian Chen, Pengpeng Xu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8198118/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 The densely distributed, scale-varying objects in unmanned aerial vehicle (UAV) images, together with dynamic, diverse, and unconstrained backgrounds, make conventional detection methods prone to missed detections, false alarms, and localization biases. To empower UAV vision tasks, we propose AD-YOLO, a unified framework tailored for traffic dense and small objects in drone imagery. First, a module, combining adaptive rotation convolution unit and grouped directional attention with mixed-kernel, is introduced to enhance the orientation invariance and multi-scale discrimination. Then, a dual-path collaborative feature pyramid network is proposed to jointly refine semantic and spatial details via multi-directional context aggregation and hierarchical spatial preservation flows. Last, a hierarchically dense reparameterized large-kernel module is designed to achieve broader receptive fields with reduced computational complexity. Extensive experiments on the VisDrone2019 and UAVDT datasets demonstrate that AD-YOLO outperforms state-of-the-art methods in detection accuracy while maintaining favorable computational efficiency, highlighting strong potential of real-life traffic monitoring applications. UAV images object detection adaptive rotation convolution feature pyramid network large kernel convolution Full Text Additional Declarations No competing interests reported. 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-8198118","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":551021732,"identity":"27f1bdf9-d42e-4498-8863-fe69d3c1e0e1","order_by":0,"name":"Yu Deng","email":"","orcid":"","institution":"South China University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Deng","suffix":""},{"id":551021739,"identity":"3e74590f-12f2-4cdd-b624-b46204bb3ccd","order_by":1,"name":"Yucong Hu","email":"","orcid":"","institution":"South China University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Yucong","middleName":"","lastName":"Hu","suffix":""},{"id":551021741,"identity":"36e428fc-5788-4d49-9666-eb22ca61cc1b","order_by":2,"name":"Yun Ye","email":"","orcid":"","institution":"Ningbo University","correspondingAuthor":false,"prefix":"","firstName":"Yun","middleName":"","lastName":"Ye","suffix":""},{"id":551021742,"identity":"353cbab7-c562-4571-9f0e-8cb520a717e4","order_by":3,"name":"Tiantian Chen","email":"","orcid":"","institution":"Korea Advanced Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Tiantian","middleName":"","lastName":"Chen","suffix":""},{"id":551021746,"identity":"b270049e-b5d2-49e3-bfb9-5fce862e4c92","order_by":4,"name":"Pengpeng Xu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6UlEQVRIiWNgGAWjYHAD5gNA4gBJWtgSoFqYidbCY0CcFnP23sOvCyru2G24kfNN4kPNHaBIP37XWfacS7OeceZZ8swZudskZxx7BhQ5jN8Wgxs5Zsa8bYeT+SVyt0nzsB0GiiQT0HL/DVDLv8PJbBI5z6T//ANquf+YkC08xo95Gw7b8UvksEkztoFsIeB9y54cM+YZxw4nSPY8M7bs7TvMY3Am2QCvFnP2M8afC2oO2xscT35448e3w3IGxw8+wO8wYBRKA+nEBgYGFgkggwe/qyBamD8DaXsgZv5AUPkoGAWjYBSMSAAAZA5MiMYLO/UAAAAASUVORK5CYII=","orcid":"","institution":"Central South University","correspondingAuthor":true,"prefix":"","firstName":"Pengpeng","middleName":"","lastName":"Xu","suffix":""}],"badges":[],"createdAt":"2025-11-25 02:53:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8198118/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8198118/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97317130,"identity":"419c3ee8-12b3-4553-aa27-f8687a298912","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6463,"visible":true,"origin":"","legend":"","description":"","filename":"4237a3e2e21e4c2aa67734ca5283e684.json","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/19d68a9d933a46172d2928bd.json"},{"id":97370787,"identity":"75f5ca5b-a248-4e50-b97c-c3f05c7b7a99","added_by":"auto","created_at":"2025-12-03 16:27:55","extension":"xml","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":112156,"visible":true,"origin":"","legend":"","description":"","filename":"4237a3e2e21e4c2aa67734ca5283e6841enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/4c961bc674a6058698d332ca.xml"},{"id":97317131,"identity":"e9a2bd70-caa4-4cee-ae8a-0050899f1ac3","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47449,"visible":true,"origin":"","legend":"","description":"","filename":"0Coverletter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/9ed7a9dc9698845f04627321.pdf"},{"id":97370436,"identity":"d7459a9c-c1e3-4fa9-9c0f-3c9207451dd5","added_by":"auto","created_at":"2025-12-03 16:27:21","extension":"jpg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":685379,"visible":true,"origin":"","legend":"","description":"","filename":"ADYOLO.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/a9d54e04f1f71d7eb1ad3a7e.jpg"},{"id":97369189,"identity":"81b15833-022c-4197-8da6-c31aab07e35b","added_by":"auto","created_at":"2025-12-03 16:23:50","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8387377,"visible":true,"origin":"","legend":"","description":"","filename":"ADYOLOUAVtrafficobjectdetection.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/1e8dcdfa28f7d6a55dfabf3d.pdf"},{"id":97317138,"identity":"7ebdf39e-5f43-42f4-9791-4c0d3ff8f5f0","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"jpg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":591143,"visible":true,"origin":"","legend":"","description":"","filename":"AG.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/935428db11ae816d32d9f6ba.jpg"},{"id":97317132,"identity":"1d2d683c-da26-408b-bc29-ab24d2632bd1","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"jpg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117050,"visible":true,"origin":"","legend":"","description":"","filename":"DPCFPN.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/a932400de5ab507d4cb85c53.jpg"},{"id":97370408,"identity":"d236ee97-717b-4b45-b0d0-7df82ee01e5d","added_by":"auto","created_at":"2025-12-03 16:27:17","extension":"jpg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":434116,"visible":true,"origin":"","legend":"","description":"","filename":"HDRepLK.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/96c6ab9801d0b13c363a18c2.jpg"},{"id":97317142,"identity":"27bc81dc-dd90-4d76-b1a1-54e6e1168481","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"jpg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1656991,"visible":true,"origin":"","legend":"","description":"","filename":"Visualizationoftheoverallattention.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/b0c091685dbbd4e8d24da549.jpg"},{"id":97370099,"identity":"756cf04a-793c-4592-80ff-c74fde52e535","added_by":"auto","created_at":"2025-12-03 16:26:44","extension":"jpg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2431779,"visible":true,"origin":"","legend":"","description":"","filename":"detectionresults.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/d6b1a13e0a13ab1455f27f7e.jpg"},{"id":97370331,"identity":"2ce05555-185c-4bfa-b5b6-a3889dbb36fe","added_by":"auto","created_at":"2025-12-03 16:27:10","extension":"eps","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2890,"visible":true,"origin":"","legend":"","description":"","filename":"empty.eps","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/36b8115c63e454c09a85c606.eps"},{"id":97317152,"identity":"3f8392c5-a38f-4712-9569-271f6e11732c","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"eps","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":91593,"visible":true,"origin":"","legend":"","description":"","filename":"fig.eps","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/4fc3b75bf9b50b3a51be44d8.eps"},{"id":97317144,"identity":"c6cc75c0-61d9-46d3-97eb-c7b2cbafa8c1","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"bst","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":146013,"visible":true,"origin":"","legend":"","description":"","filename":"snapacite.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/9129216dbb02a3a6c029f23c.bst"},{"id":97317156,"identity":"67ceeccf-cc77-434b-9f9b-e2765bc4579d","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"bst","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29828,"visible":true,"origin":"","legend":"","description":"","filename":"snaps.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/03372b6fcaacbefff5644c2e.bst"},{"id":97317153,"identity":"63a326f3-3db8-4616-9415-972ea9e9c63e","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"pdf","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":421391,"visible":true,"origin":"","legend":"","description":"","filename":"snarticle.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/efcaaee44140c5c8f295dea8.pdf"},{"id":97369292,"identity":"c40e3113-b641-46d1-b428-462ef305294e","added_by":"auto","created_at":"2025-12-03 16:24:11","extension":"bst","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35515,"visible":true,"origin":"","legend":"","description":"","filename":"snbasic.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/3b083f13fbd0b9f1989e02af.bst"},{"id":97369533,"identity":"01da1d6f-eff5-446a-9830-85eec6d30996","added_by":"auto","created_at":"2025-12-03 16:25:06","extension":"bst","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33968,"visible":true,"origin":"","legend":"","description":"","filename":"snchicago.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/57ef909047c4b85d5ab00960.bst"},{"id":97317149,"identity":"dc3bdcc0-4bae-4665-8dd1-06693ae8ca41","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"cls","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":55906,"visible":true,"origin":"","legend":"","description":"","filename":"snjnl.cls","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/ee3bd792294d322bed075b9a.cls"},{"id":97317148,"identity":"eb05b0e2-e567-4347-94c4-7fef24de4778","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"bst","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64023,"visible":true,"origin":"","legend":"","description":"","filename":"snmathphysay.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/51e0688773a586b9665f8b4f.bst"},{"id":97369602,"identity":"46d68ebf-bd4c-4601-9610-3bf7cc250ff9","added_by":"auto","created_at":"2025-12-03 16:25:18","extension":"bst","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64166,"visible":true,"origin":"","legend":"","description":"","filename":"snmathphysnum.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/a8619f11f54dc5d261a16c52.bst"},{"id":97317151,"identity":"e6399552-810e-4b7e-a6ed-3d0734c379f0","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"bst","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37333,"visible":true,"origin":"","legend":"","description":"","filename":"snnature.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/ae2f7c7cc11b2d19d38d7cfa.bst"},{"id":97370885,"identity":"a7bb0520-1c62-4a6f-b12e-65cc542785d6","added_by":"auto","created_at":"2025-12-03 16:28:06","extension":"bst","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39951,"visible":true,"origin":"","legend":"","description":"","filename":"snvancouveray.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/303e31de5d493327e8e56747.bst"},{"id":97317141,"identity":"56435445-a97f-4945-8fd2-673e834edf2d","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"bst","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40758,"visible":true,"origin":"","legend":"","description":"","filename":"snvancouvernum.bst","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/93d373454e230454bea0733a.bst"},{"id":97317145,"identity":"82d0b8f4-d01f-4f4a-875c-4be5b637ed99","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"pdf","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":418495,"visible":true,"origin":"","legend":"","description":"","filename":"usermanual.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/98ed166fafe018506b995706.pdf"},{"id":97317155,"identity":"be25da6a-77fc-432e-96de-6dbf492a287f","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"jpg","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2094219,"visible":true,"origin":"","legend":"","description":"","filename":"visualizationoftargetattention.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/598128193b4b23f4782b800c.jpg"},{"id":97317143,"identity":"a0446f65-9a60-4c12-85ab-88342f0c4220","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"xml","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":119707,"visible":true,"origin":"","legend":"","description":"","filename":"4237a3e2e21e4c2aa67734ca5283e6841structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/81af492136426552a4e86f85.xml"},{"id":97317154,"identity":"43e2afab-4153-431a-a552-4c0da313cc53","added_by":"auto","created_at":"2025-12-03 06:57:24","extension":"html","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":135734,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1/533357236c50a4ea377e038e.html"},{"id":101203750,"identity":"84587cbe-bf1c-4d19-9584-d84a8144d406","added_by":"auto","created_at":"2026-01-27 09:40:35","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8361359,"visible":true,"origin":"","legend":"","description":"","filename":"ADYOLOUAVtrafficobjectdetection.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8198118/v1_covered_bbf702e5-7697-48c4-8aee-15c96852b7c7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"AD-YOLO: A unified method for traffic dense and small object detection in UAV images","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":"UAV images, object detection, adaptive rotation convolution, feature pyramid network, large kernel convolution","lastPublishedDoi":"10.21203/rs.3.rs-8198118/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8198118/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe densely distributed, scale-varying objects in unmanned aerial vehicle (UAV) images, together with dynamic, diverse, and unconstrained backgrounds, make conventional detection methods prone to missed detections, false alarms, and localization biases. To empower UAV vision tasks, we propose AD-YOLO, a unified framework tailored for traffic dense and small objects in drone imagery. First, a module, combining adaptive rotation convolution unit and grouped directional attention with mixed-kernel, is introduced to enhance the orientation invariance and multi-scale discrimination. Then, a dual-path collaborative feature pyramid network is proposed to jointly refine semantic and spatial details via multi-directional context aggregation and hierarchical spatial preservation flows. Last, a hierarchically dense reparameterized large-kernel module is designed to achieve broader receptive fields with reduced computational complexity. Extensive experiments on the VisDrone2019 and UAVDT datasets demonstrate that AD-YOLO outperforms state-of-the-art methods in detection accuracy while maintaining favorable computational efficiency, highlighting strong potential of real-life traffic monitoring applications.\u003c/p\u003e","manuscriptTitle":"AD-YOLO: A unified method for traffic dense and small object detection in UAV images","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-03 06:57:19","doi":"10.21203/rs.3.rs-8198118/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":"31a6cd95-2c82-4198-b8f2-78ac4db94931","owner":[],"postedDate":"December 3rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-01-23T13:40:47+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-03 06:57:19","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8198118","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8198118","identity":"rs-8198118","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}