CFA-DeepLabV3+: Cross-level Fusion and Attention Network for Lightweight Road Segmentation

CFA-DeepLabV3+: Cross-level Fusion and Attention Network for Lightweight 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 CFA-DeepLabV3+: Cross-level Fusion and Attention Network for Lightweight Road Segmentation Xin Zhang, Yan Li, Zexi Hua, XiangZhen Zhou, YuGe Pan, Hui Qiao This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9178322/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract With the rapid advancement of robotics, precise environmental perception is essential for tasks such as autonomous driving and outdoor patrols. Road segmentation provides pixel-level semantic information critical for robot navigation. However, existing algorithms face two major challenges: limited datasets for diverse scenarios and high model complexity, which hinder deployment on resource-constrained platforms. To address these issues, this paper proposes CFA-DeepLabV3+ (Cross-level Fusion and Attention DeepLabV3+), a lightweight road segmentation network. It adopts MobileNetV2 as the backbone to reduce parameters and computational cost, and introduces three complementary modules: an Enhanced ASPP (E-ASPP) for multi-scale context modeling, an Adaptive Fusion Attention Module (AFAM) to dynamically balance channel and spatial attention, and a Cross-level Feature Enhancement Module (CFEM) to fuse shallow details with deep semantics. The IDD dataset is augmented with indoor corridors, forest roads, and woodland trails to boost robustness and generalization. Experimental results demonstrate that CFA-DeepLabV3 + achieves 69.40% mIoU, outperforming state-of-the-art lightweight networks while maintaining low computational overhead, offering superior real-time performance and adaptability for mobile robots in complex environments. Physical sciences/Engineering Physical sciences/Mathematics and computing Deep learning Semantic segmentation Attention mechanism Lightweight Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 24 Apr, 2026 Reviews received at journal 23 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 18 Apr, 2026 Reviewers agreed at journal 17 Apr, 2026 Reviews received at journal 03 Apr, 2026 Reviewers agreed at journal 03 Apr, 2026 Reviewers invited by journal 02 Apr, 2026 Editor assigned by journal 01 Apr, 2026 Editor invited by journal 31 Mar, 2026 Submission checks completed at journal 26 Mar, 2026 First submitted to journal 26 Mar, 2026 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-9178322","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":617337724,"identity":"71ca5e4e-9708-4a0d-ba88-f1091ce07ae1","order_by":0,"name":"Xin Zhang","email":"","orcid":"","institution":"Zhengzhou Shengda University","correspondingAuthor":false,"prefix":"","firstName":"Xin","middleName":"","lastName":"Zhang","suffix":""},{"id":617337725,"identity":"2c9b348a-ce60-4ac0-9ec8-19beb25800ad","order_by":1,"name":"Yan Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAElEQVRIiWNgGAWjYBACPmYwdYCBH8JnA5MS+LSwwbRINkD4EoS1MEC1GBxgQFiAXws7j+Hngj93EjdfO2PA+LONr87gAPPB2zwMdnm4HcZjLD2D51nitts5Bsy8bWwSBgfYkq15GJKL8WgxkOaROAzRwrgNpIXHTJqH4UBiAx5bfvMYHE7cPDsH6DCwFv5vhLQAzUw4nLhBOseAgRdiCxsBLWxl1jwHDhvPuJ1WcJj3H5vkzMNsxpZzDJJxauHnP7z5Ns+fw7L9s5M3Pvxx5hg/3/HmhzfeVNjh1MLAwGEAZxxgYDjGwACOXAOc6oGA/QEyowaf0lEwCkbBKBihAAAcjk9TS+DxmAAAAABJRU5ErkJggg==","orcid":"","institution":"Southwest Jiaotong University","correspondingAuthor":true,"prefix":"","firstName":"Yan","middleName":"","lastName":"Li","suffix":""},{"id":617337726,"identity":"10645d0b-14d3-49bd-a1bb-35de4da3d575","order_by":2,"name":"Zexi Hua","email":"","orcid":"","institution":"Southwest Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Zexi","middleName":"","lastName":"Hua","suffix":""},{"id":617337727,"identity":"3a7f63ad-5f81-4487-b613-4e1978e37798","order_by":3,"name":"XiangZhen Zhou","email":"","orcid":"","institution":"Zhengzhou Shengda University","correspondingAuthor":false,"prefix":"","firstName":"XiangZhen","middleName":"","lastName":"Zhou","suffix":""},{"id":617337728,"identity":"4c791011-002f-4e26-8e54-77532d966a18","order_by":4,"name":"YuGe Pan","email":"","orcid":"","institution":"Northwest A\u0026F University","correspondingAuthor":false,"prefix":"","firstName":"YuGe","middleName":"","lastName":"Pan","suffix":""},{"id":617337729,"identity":"baaec6a8-5929-4108-850d-f857fe6c30d0","order_by":5,"name":"Hui Qiao","email":"","orcid":"","institution":"SINOMACH Industrial Internet Research Institute (Henan) Co., Ltd","correspondingAuthor":false,"prefix":"","firstName":"Hui","middleName":"","lastName":"Qiao","suffix":""}],"badges":[],"createdAt":"2026-03-20 11:23:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9178322/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9178322/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106403843,"identity":"ae4ace9d-5984-414b-8feb-fa9c33bb0c92","added_by":"auto","created_at":"2026-04-08 09:15:03","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":906850,"visible":true,"origin":"","legend":"","description":"","filename":"CFADeepLabV3CrosslevelFusionandAttentionNetworkforLightweightRoadSegmentation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9178322/v1_covered_89566f12-b16e-4c10-ae24-9c5209c113c8.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"CFA-DeepLabV3+: Cross-level Fusion and Attention Network for Lightweight Road Segmentation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"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":"Deep learning, Semantic segmentation, Attention mechanism, Lightweight","lastPublishedDoi":"10.21203/rs.3.rs-9178322/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9178322/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWith the rapid advancement of robotics, precise environmental perception is essential for tasks such as autonomous driving and outdoor patrols. Road segmentation provides pixel-level semantic information critical for robot navigation. However, existing algorithms face two major challenges: limited datasets for diverse scenarios and high model complexity, which hinder deployment on resource-constrained platforms. To address these issues, this paper proposes CFA-DeepLabV3+ (Cross-level Fusion and Attention DeepLabV3+), a lightweight road segmentation network. It adopts MobileNetV2 as the backbone to reduce parameters and computational cost, and introduces three complementary modules: an Enhanced ASPP (E-ASPP) for multi-scale context modeling, an Adaptive Fusion Attention Module (AFAM) to dynamically balance channel and spatial attention, and a Cross-level Feature Enhancement Module (CFEM) to fuse shallow details with deep semantics. The IDD dataset is augmented with indoor corridors, forest roads, and woodland trails to boost robustness and generalization. Experimental results demonstrate that CFA-DeepLabV3\u0026thinsp;+\u0026thinsp;achieves 69.40% mIoU, outperforming state-of-the-art lightweight networks while maintaining low computational overhead, offering superior real-time performance and adaptability for mobile robots in complex environments.\u003c/p\u003e","manuscriptTitle":"CFA-DeepLabV3+: Cross-level Fusion and Attention Network for Lightweight Road Segmentation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-08 06:05:37","doi":"10.21203/rs.3.rs-9178322/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-24T07:35:30+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-23T08:40:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"256582268700998370208922460311595485788","date":"2026-04-20T07:13:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"286905479528256060537117921430153134056","date":"2026-04-18T05:55:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"215820585199895424217497921056555922426","date":"2026-04-18T03:23:24+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-03T21:37:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"335209449616201149325828095407036733015","date":"2026-04-03T12:33:26+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-02T14:32:57+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-01T06:45:08+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-31T06:25:38+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-26T12:45:07+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-03-26T12:38:17+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":"c9a9028d-5bc2-47d0-b042-2401875c14e8","owner":[],"postedDate":"April 8th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":65698862,"name":"Physical sciences/Engineering"},{"id":65698863,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-14T02:23:30+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-08 06:05:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9178322","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9178322","identity":"rs-9178322","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}