Multi-scale Feature Enhancement Pyramid,Enhance the detection capability for small targets.

Multi-scale Feature Enhancement Pyramid,Enhance the detection capability for small targets. | 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 Multi-scale Feature Enhancement Pyramid,Enhance the detection capability for small targets. jin yu shi, hao chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5279484/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 Underwater object detection has emerged as a pivotal technological challenge, playing an indispensable role in various application fields such as underwater structure maintenance, marine ecological protection, and ocean engineering projects. However, in practical application scenarios, underwater images are often complex and variable, with target objects typically being small in size. In light of this, this paper optimizes the general lightweight object detection model RT-DETR, addressing the issue of poor detection performance for small targets on traditional detection layers P3, P4, and P5. Traditional methods tend to add a P2 detection layer to enhance the detection capability for small targets, but this also introduces a series of problems, such as increased computational burden and longer post-processing time. Therefore, there is an urgent need to develop an effective feature pyramid structure specifically tailored for small targets. Based on the original Continuous Convolutional Feature Fusion (CCFF), this paper proposes an improved Multi-scale Feature Enhancement Pyramid (MFEP). Unlike the traditional approach of adding a P2 detection layer, we utilize the P2 feature layer to extract features rich in small target information through Spatial Pyramid Depthwise Convolution (SPDConv) and fuse them with the P3 and P4 layers. Subsequently, the fused feature maps undergo processing by a meticulously designed Small Object Enhancement Module. Experimental results on the DUO and RUOD datasets demonstrate that the proposed model exhibits excellent performance in underwater small object detection. Object Detection Underwater image DETR Feature Pyramid 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-5279484","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":367212697,"identity":"33bb3dbb-90cc-482c-a722-99b75421b19c","order_by":0,"name":"jin yu shi","email":"","orcid":"","institution":"Dalian Maritime University","correspondingAuthor":false,"prefix":"","firstName":"jin","middleName":"yu","lastName":"shi","suffix":""},{"id":367212698,"identity":"456641d1-7c73-434a-933c-34afb0cccf53","order_by":1,"name":"hao chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIie3PvWoCQRDA8V0Gxmb02klzeYWBgyBE8FXuCJjmirxBIgHTnFjnLRKEw3JlIGkE2yu9xiqFdgop4kdpWE2XYv/FFAs/ZtaYUOgfhgDOrTYcdxUcLw5vqZ+0GpjVr0UnMZ+YcnoJiSNKEsJe9jQnuYwgmBsmUtt/pnWV5WqiRi5mO/GSHnNbAaA5bmelmqviS+xw5iUfLKSI0Cx5T6TKBezAR+yAU1QioOWBdM8TAHG4Ow4Ij1v4LEFb94uOCGCy+8s98Wz5MB16yPVovtLvDT++RVpX6/I2jl7u3hdbDzmJ9sP9AYRCoVDol34AeyNKjhF7Du4AAAAASUVORK5CYII=","orcid":"","institution":"Dalian Maritime University","correspondingAuthor":true,"prefix":"","firstName":"hao","middleName":"","lastName":"chen","suffix":""}],"badges":[],"createdAt":"2024-10-17 04:08:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5279484/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5279484/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":67432786,"identity":"65d5d8f9-31df-48a4-aa05-4122f1d6a749","added_by":"auto","created_at":"2024-10-25 02:38:51","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2121224,"visible":true,"origin":"","legend":"","description":"","filename":"chenahopdflatex.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5279484/v1_covered_3714c481-9e30-4380-9ddf-37f6a01fe73b.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Multi-scale Feature Enhancement Pyramid,Enhance the detection capability for small targets.","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"Object Detection, Underwater image, DETR, Feature Pyramid","lastPublishedDoi":"10.21203/rs.3.rs-5279484/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5279484/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Underwater object detection has emerged as a pivotal technological challenge, playing an indispensable role in various application fields such as underwater structure maintenance, marine ecological protection, and ocean engineering projects. However, in practical application scenarios, underwater images are often complex and variable, with target objects typically being small in size. In light of this, this paper optimizes the general lightweight object detection model RT-DETR, addressing the issue of poor detection performance for small targets on traditional detection layers P3, P4, and P5. Traditional methods tend to add a P2 detection layer to enhance the detection capability for small targets, but this also introduces a series of problems, such as increased computational burden and longer post-processing time. Therefore, there is an urgent need to develop an effective feature pyramid structure specifically tailored for small targets. Based on the original Continuous Convolutional Feature Fusion (CCFF), this paper proposes an improved Multi-scale Feature Enhancement Pyramid (MFEP). Unlike the traditional approach of adding a P2 detection layer, we utilize the P2 feature layer to extract features rich in small target information through Spatial Pyramid Depthwise Convolution (SPDConv) and fuse them with the P3 and P4 layers. Subsequently, the fused feature maps undergo processing by a meticulously designed Small Object Enhancement Module. Experimental results on the DUO and RUOD datasets demonstrate that the proposed model exhibits excellent performance in underwater small object detection.","manuscriptTitle":"Multi-scale Feature Enhancement Pyramid,Enhance the detection capability for small targets.","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-21 13:26:36","doi":"10.21203/rs.3.rs-5279484/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":"50503be5-fd69-4595-b1d4-60dd96eaa9c2","owner":[],"postedDate":"October 21st, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-10-25T02:38:31+00:00","versionOfRecord":[],"versionCreatedAt":"2024-10-21 13:26:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5279484","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5279484","identity":"rs-5279484","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}