DTRoadseg: A duplex transform heterogeneous feature fusion network for road segmentation

DTRoadseg: A duplex transform heterogeneous feature fusion network for road segmentation | 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 Article DTRoadseg: A duplex transform heterogeneous feature fusion network for road segmentation Zhiyang Guo, Xing Hu, Jiejia Wang, XiaoYu Miao, MengTeng Sun, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3912966/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 29 Jul, 2024 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract Detecting roads in automatic driving environments poses a challenge due to issues such as boundary fuzziness, occlusion, and glare from light. We believe that two factors are instrumental in addressing these challenges and enhancing detection performance: global context dependency and effective feature representation that prioritizes important feature channels. To tackle these issues, we introduce DTRoadseg, a novel duplex Transformer-based heterogeneous feature fusion network designed for road segmentation. DTRoadseg leverages a duplex encoder architecture to extract heterogeneous features from both RGB images and point-cloud depth images. Subsequently, we introduce a multi-source Heterogeneous Feature Reinforcement Block (HFRB) for fusion of the encoded features, comprising a Heterogeneous Feature Fusion Module (HFFM) and a Reinforcement Fusion Module (RFM). The HFFM leverages the self-attention mechanisms of Transformers to achieve effective fusion through token interactions, while the RFM focuses on emphasizing informative features while downplaying less important ones, thereby reinforcing feature fusion. Finally, a Transformer decoder is utilized to produce the final semantic prediction. Furthermore, we employ a boundary loss function to optimize the segmentation structure area, reduce false detection areas, and improve model accuracy. Extensive experiments are carried out on the KITTI road dataset. The results demonstrate that, compared with state-of-the-art methods, DTRoadseg exhibits superior performance in terms of segmentation accuracy. Physical sciences/Engineering Physical sciences/Mathematics and computing/Information technology road segmentation heterogeneous feature feature fusion Transformer attention mechanism Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 29 Jul, 2024 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 26 Mar, 2024 Reviews received at journal 25 Mar, 2024 Reviews received at journal 12 Mar, 2024 Reviews received at journal 11 Mar, 2024 Reviewers agreed at journal 28 Feb, 2024 Reviewers agreed at journal 27 Feb, 2024 Reviewers invited by journal 27 Feb, 2024 Editor assigned by journal 24 Feb, 2024 Editor invited by journal 21 Feb, 2024 Submission checks completed at journal 21 Feb, 2024 First submitted to journal 31 Jan, 2024 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-3912966","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":274135497,"identity":"0734bffa-59aa-45a5-b601-913541a8b2f6","order_by":0,"name":"Zhiyang Guo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1klEQVRIiWNgGAWjYBAC+8OMjQ8+GPxn5mdvAHJriNFznPmw4YwKZnbJngNA3jFitJxnS5PmOcPMbzAjgYGBsYkIHYzNPGaSM9vYpA0kH2+8wdhgE01QCzMzj7HFxzYeY3PptGILxoa03AZCWtiYeQxvzmyTSLacnWMmwdhwmLAWHmYeA2neNoP6DTfPEKlFgpktCej9BGaDGzxALU1EaDFgBgfyAWbJHqBfEo4R4RcD/oOgqDwAjMrDG298qLEhrAVFu0QCKcohWkjVMQpGwSgYBSMDAAA4VD2W6EnnZwAAAABJRU5ErkJggg==","orcid":"","institution":"Jiangsu Shipping College","correspondingAuthor":true,"prefix":"","firstName":"Zhiyang","middleName":"","lastName":"Guo","suffix":""},{"id":274135498,"identity":"12d33fc2-ca6e-4be6-9c4b-2947b119f814","order_by":1,"name":"Xing Hu","email":"","orcid":"","institution":"University of Shanghai for Science \u0026 Technology","correspondingAuthor":false,"prefix":"","firstName":"Xing","middleName":"","lastName":"Hu","suffix":""},{"id":274135499,"identity":"876351fd-0800-4cde-ac63-9e5329192365","order_by":2,"name":"Jiejia Wang","email":"","orcid":"","institution":"Jiangsu Shipping College","correspondingAuthor":false,"prefix":"","firstName":"Jiejia","middleName":"","lastName":"Wang","suffix":""},{"id":274135500,"identity":"c3af2940-ba85-4f85-aec5-2485b8049a07","order_by":3,"name":"XiaoYu Miao","email":"","orcid":"","institution":"Jiangsu Shipping College","correspondingAuthor":false,"prefix":"","firstName":"XiaoYu","middleName":"","lastName":"Miao","suffix":""},{"id":274135501,"identity":"d47d43bd-ce97-4c32-b8cf-b033d494077b","order_by":4,"name":"MengTeng Sun","email":"","orcid":"","institution":"Jiangsu Shipping College","correspondingAuthor":false,"prefix":"","firstName":"MengTeng","middleName":"","lastName":"Sun","suffix":""},{"id":274135502,"identity":"80447d8b-ba39-4374-8742-0b741ac4e566","order_by":5,"name":"HuaiWei Wang","email":"","orcid":"","institution":"Jiangsu Shipping College","correspondingAuthor":false,"prefix":"","firstName":"HuaiWei","middleName":"","lastName":"Wang","suffix":""},{"id":274135503,"identity":"6f5de64b-26f6-4e0e-9c6d-4a670d57d9a4","order_by":6,"name":"XueYing Ma","email":"","orcid":"","institution":"Jiangsu Shipping College","correspondingAuthor":false,"prefix":"","firstName":"XueYing","middleName":"","lastName":"Ma","suffix":""}],"badges":[],"createdAt":"2024-01-31 07:30:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3912966/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3912966/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-024-68255-4","type":"published","date":"2024-07-29T15:57:16+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":61793376,"identity":"022e02f3-dcc2-4e17-b086-d2e19e064dd6","added_by":"auto","created_at":"2024-08-05 16:11:43","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":889763,"visible":true,"origin":"","legend":"","description":"","filename":"DTRoadsegAduplextransformheterogeneousfeaturefusionnetworkforroadsegmentation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3912966/v1_covered_04c5f5e1-d752-4a55-a3c3-ede0e8b8d188.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"DTRoadseg: A duplex transform heterogeneous feature fusion network for road segmentation","fulltext":[],"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":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"road segmentation, heterogeneous feature, feature fusion, Transformer, attention mechanism","lastPublishedDoi":"10.21203/rs.3.rs-3912966/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3912966/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDetecting roads in automatic driving environments poses a challenge due to issues such as boundary fuzziness, occlusion, and glare from light. We believe that two factors are instrumental in addressing these challenges and enhancing detection performance: global context dependency and effective feature representation that prioritizes important feature channels. To tackle these issues, we introduce DTRoadseg, a novel duplex Transformer-based heterogeneous feature fusion network designed for road segmentation. DTRoadseg leverages a duplex encoder architecture to extract heterogeneous features from both RGB images and point-cloud depth images. Subsequently, we introduce a multi-source Heterogeneous Feature Reinforcement Block (HFRB) for fusion of the encoded features, comprising a Heterogeneous Feature Fusion Module (HFFM) and a Reinforcement Fusion Module (RFM). The HFFM leverages the self-attention mechanisms of Transformers to achieve effective fusion through token interactions, while the RFM focuses on emphasizing informative features while downplaying less important ones, thereby reinforcing feature fusion. Finally, a Transformer decoder is utilized to produce the final semantic prediction. Furthermore, we employ a boundary loss function to optimize the segmentation structure area, reduce false detection areas, and improve model accuracy. Extensive experiments are carried out on the KITTI road dataset. The results demonstrate that, compared with state-of-the-art methods, DTRoadseg exhibits superior performance in terms of segmentation accuracy.\u003c/p\u003e","manuscriptTitle":"DTRoadseg: A duplex transform heterogeneous feature fusion network for road segmentation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-02-23 20:37:11","doi":"10.21203/rs.3.rs-3912966/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-03-26T06:08:05+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-03-25T14:23:21+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-03-12T13:37:50+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-03-12T00:34:21+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"6173008b-04d7-4897-abfe-6778f0f21c78","date":"2024-02-29T02:32:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"e802face-95c4-4d68-a83e-36b10853d267","date":"2024-02-27T17:15:25+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-02-27T09:29:31+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-02-24T14:40:52+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-02-21T10:19:29+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-02-21T10:09:59+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-01-31T07:20:22+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"68af61e6-8140-4f75-8c26-ac55fc6fffdf","owner":[],"postedDate":"February 23rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":28895983,"name":"Physical sciences/Engineering"},{"id":28895984,"name":"Physical sciences/Mathematics and computing/Information technology"}],"tags":[],"updatedAt":"2024-08-05T16:01:28+00:00","versionOfRecord":{"articleIdentity":"rs-3912966","link":"https://doi.org/10.1038/s41598-024-68255-4","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2024-07-29 15:57:16","publishedOnDateReadable":"July 29th, 2024"},"versionCreatedAt":"2024-02-23 20:37:11","video":"","vorDoi":"10.1038/s41598-024-68255-4","vorDoiUrl":"https://doi.org/10.1038/s41598-024-68255-4","workflowStages":[]},"version":"v1","identity":"rs-3912966","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3912966","identity":"rs-3912966","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}