Traffic sign detection algorithm with occlusion awareness and edge enhancement

Traffic sign detection algorithm with occlusion awareness and edge enhancement | 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 Traffic sign detection algorithm with occlusion awareness and edge enhancement Jun Li, Weijuan Shi, Binbin Ding, Lin Yang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6407450/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 long-distance small-size traffic signs and the low recognitionaccuracy in scenarios with background interference and occlusion, we propose the MSED-YOLOv8detection framework. The MFE (Multi-scale Feature Enhancement) module employs dynamic upsamplingand hybrid pooling to align multi-scale features and mitigate feature information loss. TheSMSA (Shielding-aware Multi-Scale Attention) mechanism filters important features through twodimensionalspatial-channel screening and integrates them into the network head to enhance featureinformation and improve robustness in occlusion scenarios. The ESF (Edge-Spatial informationFusion) module realizes edge enhancement and spatial awareness to suppress background interference.The DSFF (Dynamic Sequence Feature Fusion) module efficiently integrates deep semantic featuresand shallow detail information to strengthen the detection capability for small targets. Experimentalresults demonstrate that the MSED-YOLOv8 algorithm achieves average accuracy improvementsof 7.9%, 4.2%, and 2.9% on the TT100k, CCTSDB, and GTSDB datasets, respectively, comparedwith the YOLOv8s benchmark model. Additionally, the number of parameters and the model sizeare reduced by 17.1% and 8.0%, respectively. The proposed algorithm not only enhances detectionperformance but also achieves model lightweighting, making it more suitable for traffic sign detectionin complex traffic scenes. traffic sign detection YOLOv8 occlusion scene lightweighting feature enhancement object detection 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-6407450","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":443077286,"identity":"9edad51d-e40c-4a0a-b706-002639dd3a53","order_by":0,"name":"Jun Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIiWNgGAWjYBACPmaGBCBlw8DAzNzA8ICBjbAWNoiWNKAWxgYgk02CsBYIdRiIwVoYiNDCzvDwc0HF+Wj+dpCWP3x1BgeYnz0g4LBk6RlnbufOOAzSwsMmYXCAzdyAkF+kedtu5zaAtUiAtPDg9xDIlt+8/87lzgdrMSBOS5o0b8OB3A1gLQlEarHmOZacuxGo5UDCATbJmYfZzPBq4ec/k3ybp8Yud975wwcffPhzjJ/vePMzAiHNkwBnHmBgOAaMU/zqgYD9ADKvhqD6UTAKRsEoGHkAAJYZQRTb7W4sAAAAAElFTkSuQmCC","orcid":"","institution":"Lanzhou Jiaotong University","correspondingAuthor":true,"prefix":"","firstName":"Jun","middleName":"","lastName":"Li","suffix":""},{"id":443077287,"identity":"f23f6017-5b7d-4862-9e58-2c838cdb3420","order_by":1,"name":"Weijuan Shi","email":"","orcid":"","institution":"Lanzhou Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Weijuan","middleName":"","lastName":"Shi","suffix":""},{"id":443077288,"identity":"214eabc6-f57a-415b-b655-35eeb2d898c7","order_by":2,"name":"Binbin Ding","email":"","orcid":"","institution":"Lanzhou Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Binbin","middleName":"","lastName":"Ding","suffix":""},{"id":443077289,"identity":"a4f0b6ba-4f96-4475-898d-eefadaa6228a","order_by":3,"name":"Lin Yang","email":"","orcid":"","institution":"Lanzhou Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Lin","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2025-04-09 03:08:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6407450/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6407450/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":81667667,"identity":"4dee45e7-0ed0-4979-ab35-f62218f0a242","added_by":"auto","created_at":"2025-04-30 03:59:11","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2276221,"visible":true,"origin":"","legend":"","description":"","filename":"Trafficsigndetectionalgorithmwithocclusionawarenessandedgeenhancement.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6407450/v1_covered_9083aec3-3263-411d-b511-fa399745a04c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Traffic sign detection algorithm with occlusion awareness and edge enhancement","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":"traffic sign detection, YOLOv8, occlusion scene, lightweighting, feature enhancement, object detection","lastPublishedDoi":"10.21203/rs.3.rs-6407450/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6407450/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"To address the challenges of detecting long-distance small-size traffic signs and the low recognitionaccuracy in scenarios with background interference and occlusion, we propose the MSED-YOLOv8detection framework. The MFE (Multi-scale Feature Enhancement) module employs dynamic upsamplingand hybrid pooling to align multi-scale features and mitigate feature information loss. TheSMSA (Shielding-aware Multi-Scale Attention) mechanism filters important features through twodimensionalspatial-channel screening and integrates them into the network head to enhance featureinformation and improve robustness in occlusion scenarios. The ESF (Edge-Spatial informationFusion) module realizes edge enhancement and spatial awareness to suppress background interference.The DSFF (Dynamic Sequence Feature Fusion) module efficiently integrates deep semantic featuresand shallow detail information to strengthen the detection capability for small targets. Experimentalresults demonstrate that the MSED-YOLOv8 algorithm achieves average accuracy improvementsof 7.9%, 4.2%, and 2.9% on the TT100k, CCTSDB, and GTSDB datasets, respectively, comparedwith the YOLOv8s benchmark model. Additionally, the number of parameters and the model sizeare reduced by 17.1% and 8.0%, respectively. The proposed algorithm not only enhances detectionperformance but also achieves model lightweighting, making it more suitable for traffic sign detectionin complex traffic scenes.","manuscriptTitle":"Traffic sign detection algorithm with occlusion awareness and edge enhancement","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-30 03:51:06","doi":"10.21203/rs.3.rs-6407450/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":"b28df41c-ea38-48ce-a14c-6b01c0895d6a","owner":[],"postedDate":"April 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-04-30T03:51:06+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-30 03:51:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6407450","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6407450","identity":"rs-6407450","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}