Dim Target Detection Method Based on Deep Learning in Complex Traffic Environment

Dim Target Detection Method Based on Deep Learning in Complex Traffic Environment | 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 Dim Target Detection Method Based on Deep Learning in Complex Traffic Environment Hao Zheng, Jianfang Liu, Xiaogang Ren This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-177944/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 01 Mar, 2022 Read the published version in Journal of Grid Computing → Version 1 posted You are reading this latest preprint version Abstract Although the current vehicle detection and recognition framework based on deep learning has its own characteristics and advantages, it is difficult to effectively combine multi-scale and multi category vehicle features, and there is still room for improvement in vehicle detection and recognition performance. Based on this, an improved fast R-CNN convolutional neural network is proposed to detect dim targets in complex traffic environment. The deep learning model of fast R-CNN convolutional neural network is introduced into the image recognition of complex traffic environment, and a structure optimization method is proposed, which replaces vgg16 in fast RCNN with RESNET to make it suitable for small target recognition in complex background. Max pooling is the down sampling method, and then feature pyramid network is introduced into RPN to generate target candidate box to optimize the structure of convolutional neural network. After training with 1497 images, the complex traffic environment images are identified and tested. Software Engineering Object detection Faster R-CNN Residual Network (ResNet) Deep learning Complex traffic environment Structure optimization Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Full Text Cite Share Download PDF Status: Published Journal Publication published 01 Mar, 2022 Read the published version in Journal of Grid Computing → 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-177944","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":10743944,"identity":"575b7b59-42b7-4428-830c-cc3ac4772194","order_by":0,"name":"Hao Zheng","email":"","orcid":"","institution":"Pingdingshan University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Zheng","suffix":""},{"id":10743945,"identity":"9fb9ae52-1a55-435e-9eca-34a5c61d5ef1","order_by":1,"name":"Jianfang Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIie3PPQrCMBTA8VcCyfKwaybP8CBQBItnMQidQik4OugBegCP4Q0Ugjq5uzvqkFFQ0PjRtc0omP+QvCE/kgDEYj9Yj/lFEoBY+MFR3u8kvCG4AUiWVaG6STO8CENndTcRSDCoRiWKgz3nxMYg7HbV/jBPJE2miGUxNMRLwKI4BhCm12AyZQinIDELIXNdp5dMDUjqRSCxupZGnYAohPDKk70nlyypaax411/S1K6YvM/8w4xy1/ujnwq7ayU+cZPfG987bzvblLjPzlzI6VgsFvu/nvKaOrFQkUjbAAAAAElFTkSuQmCC","orcid":"","institution":"PingDingShan University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Jianfang","middleName":"","lastName":"Liu","suffix":""},{"id":10743946,"identity":"e02b7c85-29e6-4719-9a0a-5b18dd45ec70","order_by":2,"name":"Xiaogang Ren","email":"","orcid":"","institution":"Soochow University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xiaogang","middleName":"","lastName":"Ren","suffix":""}],"badges":[],"createdAt":"2021-01-28 16:29:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-177944/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-177944/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10723-021-09594-8","type":"published","date":"2022-03-01T18:57:25+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":5792594,"identity":"5a70cefc-471c-4491-983f-dce58145559e","added_by":"auto","created_at":"2021-02-09 18:48:53","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":75094,"visible":true,"origin":"","legend":"The structure of target detection network is proposed","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1/b182af60df28f5c3d3044d62.jpg"},{"id":5792589,"identity":"5240744e-9d2b-4430-83be-593dfa191c05","added_by":"auto","created_at":"2021-02-09 18:48:53","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":77182,"visible":true,"origin":"","legend":"Region proposal networks architecture","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1/9043ba4b48ea3bc225c2d9ef.jpg"},{"id":5792734,"identity":"4e09f641-3e34-48f3-9abb-7f70a3f5b6b9","added_by":"auto","created_at":"2021-02-09 18:51:53","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":65400,"visible":true,"origin":"","legend":"Datasets used in the experiment","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1/b81842edd810cbf37ce5ace7.jpg"},{"id":5792591,"identity":"1f698d4a-274d-42cc-a3e9-eda3b0df274f","added_by":"auto","created_at":"2021-02-09 18:48:53","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":39374,"visible":true,"origin":"","legend":"Loss curves of training","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1/b7c630e8599e770f0a820d5b.jpg"},{"id":5792737,"identity":"ec215101-708d-460b-9f94-7f529412468f","added_by":"auto","created_at":"2021-02-09 18:51:53","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":91116,"visible":true,"origin":"","legend":"Recognition effect picture","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1/90c74b015e72644aec392d7b.jpg"},{"id":5792735,"identity":"86a5503b-bf65-4c7b-9897-3b4c5fec0e0c","added_by":"auto","created_at":"2021-02-09 18:51:53","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":69708,"visible":true,"origin":"","legend":"Model precision comparison","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1/1d6935270ba5fc6968ae5e6d.jpg"},{"id":5792889,"identity":"88ef9eef-3e84-47c2-a412-fed99e78b449","added_by":"auto","created_at":"2021-02-09 18:54:53","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":69376,"visible":true,"origin":"","legend":"Model recall comparison","description":"","filename":"7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1/5f8930a63e75db2ab15d2b89.jpg"},{"id":5792738,"identity":"06e5ca47-a294-4daa-8f84-b884b154405b","added_by":"auto","created_at":"2021-02-09 18:51:53","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":59725,"visible":true,"origin":"","legend":"Precision recall curve of vehicle and dim target","description":"","filename":"8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1/d5f317ae0ae3239d9b45f9ab.jpg"},{"id":18869786,"identity":"2f43dc96-ff45-4b1e-82f2-0b92d780283b","added_by":"auto","created_at":"2022-03-04 18:57:37","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":849959,"visible":true,"origin":"","legend":"","description":"","filename":"HaoZhengSoftComputingArticle.pdf","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1_covered.pdf"},{"id":13582582,"identity":"46a4aa21-c968-4968-8baa-904ddd26b350","added_by":"auto","created_at":"2021-09-17 04:31:42","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":844553,"visible":true,"origin":"","legend":"","description":"","filename":"HaoZhengSoftComputingArticle.pdf","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1_covered.pdf"},{"id":5793041,"identity":"a2ab9cec-0b9f-480f-b827-30d194cfcdd9","added_by":"auto","created_at":"2021-02-09 18:57:56","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":490480,"visible":true,"origin":"","legend":"","description":"","filename":"HaoZhengSoftComputingArticle.pdf","url":"https://assets-eu.researchsquare.com/files/rs-177944/v1_stamped.pdf"}],"financialInterests":"","formattedTitle":"\u003cp\u003eDim Target Detection Method Based on Deep Learning in Complex Traffic Environment\u003c/p\u003e","fulltext":[{"header":"Full Text","content":"\u003cp\u003eThis preprint is available for \u003ca href='/article/rs-177944/latest.pdf' target='_blank'\u003edownload as a PDF\u003c/a\u003e.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"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, Faster R-CNN, Residual Network (ResNet), Deep learning, Complex traffic environment, Structure optimization","lastPublishedDoi":"10.21203/rs.3.rs-177944/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-177944/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAlthough the current vehicle detection and recognition framework based on deep learning has its own characteristics and advantages, it is difficult to effectively combine multi-scale and multi category vehicle features, and there is still room for improvement in vehicle detection and recognition performance. Based on this, an improved fast R-CNN convolutional neural network is proposed to detect dim targets in complex traffic environment. The deep learning model of fast R-CNN convolutional neural network is introduced into the image recognition of complex traffic environment, and a structure optimization method is proposed, which replaces vgg16 in fast RCNN with RESNET to make it suitable for small target recognition in complex background. Max pooling is the down sampling method, and then feature pyramid network is introduced into RPN to generate target candidate box to optimize the structure of convolutional neural network. After training with 1497 images, the complex traffic environment images are identified and tested.\u003c/p\u003e","manuscriptTitle":"Dim Target Detection Method Based on Deep Learning in Complex Traffic Environment","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-02-09 18:48:51","doi":"10.21203/rs.3.rs-177944/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":"1894de3e-eb9f-444c-8fcb-c03d82696b61","owner":[],"postedDate":"February 9th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":2332886,"name":"Software Engineering"}],"tags":[],"updatedAt":"2022-03-04T18:57:25+00:00","versionOfRecord":{"articleIdentity":"rs-177944","link":"https://doi.org/10.1007/s10723-021-09594-8","journal":{"identity":"journal-of-grid-computing","isVorOnly":false,"title":"Journal of Grid Computing"},"publishedOn":"2022-03-01 18:57:25","publishedOnDateReadable":"March 1st, 2022"},"versionCreatedAt":"2021-02-09 18:48:51","video":"","vorDoi":"10.1007/s10723-021-09594-8","vorDoiUrl":"https://doi.org/10.1007/s10723-021-09594-8","workflowStages":[]},"version":"v1","identity":"rs-177944","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-177944","identity":"rs-177944","version":["v1"]},"buildId":"ehx78VzkSd0WSzXnipQa-","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}