Integrating Convexity and DeepLabv3+ for SemanticSegmentation of Power Lines

Integrating Convexity and DeepLabv3+ for SemanticSegmentation of Power Lines | 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 Integrating Convexity and DeepLabv3+ for SemanticSegmentation of Power Lines Ying Hou, Jindou Tuo, Haochen Li, Zhuoxin Yan, Xuemei Huang, Yuqi Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8607246/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract To address the challenges of segmenting slender, low‑contrast power lines in complex aerial inspection imagery, this paper proposes PLE‑DeepLabv3+, an enhanced deep learning model based on DeepLabv3+. The method employs a channel reconfigured ConvNeXtV2 backbone to maintain high resolution spatial details while reducing parameters, incorporates a direction selective coordinate attention module (CA_DS) to enhance the perception of linear structures, and replaces standard convolutions in the atrous spatial pyramid pooling (ASPP) block with depthwise separable convolutions for efficient multi scale fusion. Additionally, an adaptive edge detection module (AEDM) is integrated to reinforce weak boundaries, and a category aware focal loss (FL_AL) is designed to alleviate extreme class imbalance. Evaluated on a composite power line dataset, the proposed model achieves 85.2% mIoU and 92.7% mPA, with only 2.3 M parameters and 78.7 GFLOPs, surpassing both general and task specialized segmentation networks. Cross dataset tests confirm its strong generalization across varied scenes. The lightweight and accurate architecture provides a reliable visual solution for automated inspection in power grid monitoring systems. Physical sciences/Engineering Physical sciences/Mathematics and computing Power Line Extraction Semantic Segmentation DeepLabv3+ ConvNeXtv2 Adaptive Edge Enhancement Module Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 27 Feb, 2026 Reviewers agreed at journal 25 Feb, 2026 Reviewers invited by journal 19 Jan, 2026 Editor invited by journal 19 Jan, 2026 Editor assigned by journal 16 Jan, 2026 Submission checks completed at journal 16 Jan, 2026 First submitted to journal 15 Jan, 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-8607246","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":577508743,"identity":"a4fc0a56-187e-45fb-bcee-0e8f53c4e7e2","order_by":0,"name":"Ying Hou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuklEQVRIiWNgGAWjYBACNvb+5x8+/LCx4ydaCx/PGTbGmT1pyZINxGqRk8hhY+ZhO8y44QDRDpPIPfaYh+cws/Hx5A0MPyq2EaGF51264RyLdD6zM88KGHvO3CZCC3uCgcQbHmtmsxs5BsyMbcRoYQBq4WFjZtw8g2gtHDlmkjxszowbJIjWwnMs2RAUyBJAvxwkyi/y7c0HH4Cjsj1544MfFURoQQIJBgdIUg/WQqqOUTAKRsEoGCEAADyxO1cVGoN0AAAAAElFTkSuQmCC","orcid":"","institution":"Xi'an University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Ying","middleName":"","lastName":"Hou","suffix":""},{"id":577508744,"identity":"73312d84-49c6-45be-a5e6-afa93a83e9dd","order_by":1,"name":"Jindou Tuo","email":"","orcid":"","institution":"Xi'an University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Jindou","middleName":"","lastName":"Tuo","suffix":""},{"id":577508745,"identity":"ab3115de-2fd3-4ecc-9cda-aac1146b714c","order_by":2,"name":"Haochen Li","email":"","orcid":"","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Haochen","middleName":"","lastName":"Li","suffix":""},{"id":577508746,"identity":"b7ba91d5-a203-40ca-9dd1-942c46027755","order_by":3,"name":"Zhuoxin Yan","email":"","orcid":"","institution":"Xi'an University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Zhuoxin","middleName":"","lastName":"Yan","suffix":""},{"id":577508747,"identity":"8c400951-6afe-4a0f-bee3-7634be4bfb82","order_by":4,"name":"Xuemei Huang","email":"","orcid":"","institution":"Xi'an University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Xuemei","middleName":"","lastName":"Huang","suffix":""},{"id":577508750,"identity":"598bbfc2-8948-4db2-bca1-65701a95d0d2","order_by":5,"name":"Yuqi Wang","email":"","orcid":"","institution":"Xi'an University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Yuqi","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2026-01-15 05:53:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8607246/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8607246/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":100858542,"identity":"088fc884-ab50-443f-a17f-380d2cb57aac","added_by":"auto","created_at":"2026-01-22 07:24:25","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7059,"visible":true,"origin":"","legend":"","description":"","filename":"9dc3752f71684a13a5696a54b5c900c6.json","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/cb5bd10d79067b2d862a5f2c.json"},{"id":100858352,"identity":"612e0a9a-4d29-4c1c-8e98-7199b6fc8c5f","added_by":"auto","created_at":"2026-01-22 07:24:13","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":85641,"visible":true,"origin":"","legend":"","description":"","filename":"Coverletter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/d4ac0207bdef83a772cb2ef4.pdf"},{"id":100830579,"identity":"5dcae00c-f136-4058-ae58-6eb385141943","added_by":"auto","created_at":"2026-01-21 20:41:38","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":197550,"visible":true,"origin":"","legend":"","description":"","filename":"Deeplabv3ConvnextV2.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/f6f4f055a8b165be96e400f0.png"},{"id":100858694,"identity":"c0a3f841-a4f4-476c-893e-eac701b0215f","added_by":"auto","created_at":"2026-01-22 07:24:37","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2961259,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/169e6fdb113146db4883a6b1.pdf"},{"id":100858754,"identity":"26670c81-75f3-449e-82e9-5651d0bf553b","added_by":"auto","created_at":"2026-01-22 07:24:44","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":732367,"visible":true,"origin":"","legend":"","description":"","filename":"figure10.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/94b84452c0009c17188c7d47.png"},{"id":101751146,"identity":"7ec58e8a-b84b-4d89-8c0d-8a68353a513d","added_by":"auto","created_at":"2026-02-03 10:16:48","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":592637,"visible":true,"origin":"","legend":"","description":"","filename":"figure11.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/7046c08d59df32f4fed2e423.png"},{"id":100858450,"identity":"e05451a0-0d5c-44c6-9de5-041df6fe9d84","added_by":"auto","created_at":"2026-01-22 07:24:18","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":354595,"visible":true,"origin":"","legend":"","description":"","filename":"figure12.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/c2e6097896aca22a65d946f1.png"},{"id":100949422,"identity":"5433e1f7-3d6d-4081-baab-9775717d6e24","added_by":"auto","created_at":"2026-01-23 07:02:08","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":443039,"visible":true,"origin":"","legend":"","description":"","filename":"figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/c33abdde58a85cd06de94e7e.png"},{"id":100858354,"identity":"918099d0-4035-4eff-9301-d5e2d9c56af3","added_by":"auto","created_at":"2026-01-22 07:24:13","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":440888,"visible":true,"origin":"","legend":"","description":"","filename":"figure2new.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/52af32a3ccb522e49f1a4a98.png"},{"id":100858509,"identity":"d1e25bde-8f1c-4b28-b2f4-45f1d7256f58","added_by":"auto","created_at":"2026-01-22 07:24:23","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":45975,"visible":true,"origin":"","legend":"","description":"","filename":"figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/da8bcecfb58b96c6fa92912e.png"},{"id":100830574,"identity":"47c83970-fbf0-41d4-ad35-b7d6cddc132b","added_by":"auto","created_at":"2026-01-21 20:41:38","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23831,"visible":true,"origin":"","legend":"","description":"","filename":"figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/6b4341f9d57280f8b939c0ec.png"},{"id":100830573,"identity":"473ec8a7-dd74-4817-8fde-039b90623213","added_by":"auto","created_at":"2026-01-21 20:41:38","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":61614,"visible":true,"origin":"","legend":"","description":"","filename":"figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/7a4764ccce602feaeaf73eee.png"},{"id":100859190,"identity":"5480237b-fd73-41cf-a75b-77888c46a167","added_by":"auto","created_at":"2026-01-22 07:25:33","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":80640,"visible":true,"origin":"","legend":"","description":"","filename":"figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/645b49f261cb2af92547d465.png"},{"id":100859020,"identity":"e5ab6976-02d5-4dfc-9e4a-9f12527f1cd2","added_by":"auto","created_at":"2026-01-22 07:25:06","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":319357,"visible":true,"origin":"","legend":"","description":"","filename":"figure9.png","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/6876a14dc7f86350f5a433cf.png"},{"id":100858971,"identity":"719b9d73-5bb1-4d1d-a124-76e33e0b6239","added_by":"auto","created_at":"2026-01-22 07:25:02","extension":"ldf","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49107,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrvltwaall.ldf","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/4a250718bf09b0c1abfdbc8d.ldf"},{"id":100830581,"identity":"11a09004-6dea-445b-9e82-8314bbaca493","added_by":"auto","created_at":"2026-01-21 20:41:38","extension":"ldf","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":268657,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrvltwaen.ldf","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/ed19e45b76e37ea0c4bb9faf.ldf"},{"id":100859204,"identity":"2d30599a-c93d-437c-b3dc-bb5cf0df7f2b","added_by":"auto","created_at":"2026-01-22 07:25:37","extension":"sty","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":15480,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrv.sty","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/e4486135c25f8c5c726a1485.sty"},{"id":100830586,"identity":"c222e79c-c3c4-418b-bfdb-bf6e91055db0","added_by":"auto","created_at":"2026-01-21 20:41:38","extension":"bst","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36153,"visible":true,"origin":"","legend":"","description":"","filename":"naturemagdoi.bst","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/bdd34dbf7ff4c2dd92396907.bst"},{"id":100830584,"identity":"301bd451-bf32-46af-a035-68bc99978e3f","added_by":"auto","created_at":"2026-01-21 20:41:38","extension":"jpg","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":379752,"visible":true,"origin":"","legend":"","description":"","filename":"stream.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/98fdf5e496ee086a5f1f090d.jpg"},{"id":100830588,"identity":"6c1036d9-13ba-4c62-8655-2795c25840e4","added_by":"auto","created_at":"2026-01-21 20:41:38","extension":"cls","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5824,"visible":true,"origin":"","legend":"","description":"","filename":"wlscirep.cls","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/00e5cf719a5e8d97f95107c7.cls"},{"id":100858395,"identity":"0561a5df-7326-4e59-97f9-6b40da45992d","added_by":"auto","created_at":"2026-01-22 07:24:14","extension":"xml","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":65825,"visible":true,"origin":"","legend":"","description":"","filename":"9dc3752f71684a13a5696a54b5c900c61structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1/5aa2cf0284cf5b98378b02af.xml"},{"id":101755823,"identity":"c8ac7943-faf9-4146-9bf8-f0221bcb1c56","added_by":"auto","created_at":"2026-02-03 10:55:08","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":770847,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8607246/v1_covered_c22e086e-8098-481f-98d1-0e5cc432f614.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Integrating Convexity and DeepLabv3+ for SemanticSegmentation of Power Lines","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":"Power Line Extraction, Semantic Segmentation, DeepLabv3+, ConvNeXtv2, Adaptive Edge Enhancement Module","lastPublishedDoi":"10.21203/rs.3.rs-8607246/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8607246/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"To address the challenges of segmenting slender, low‑contrast power lines in complex aerial inspection imagery, this paper proposes PLE‑DeepLabv3+, an enhanced deep learning model based on DeepLabv3+. The method employs a channel reconfigured ConvNeXtV2 backbone to maintain high resolution spatial details while reducing parameters, incorporates a direction selective coordinate attention module (CA\\_DS) to enhance the perception of linear structures, and replaces standard convolutions in the atrous spatial pyramid pooling (ASPP) block with depthwise separable convolutions for efficient multi scale fusion. Additionally, an adaptive edge detection module (AEDM) is integrated to reinforce weak boundaries, and a category aware focal loss (FL\\_AL) is designed to alleviate extreme class imbalance. Evaluated on a composite power line dataset, the proposed model achieves 85.2\\% mIoU and 92.7\\% mPA, with only 2.3 M parameters and 78.7 GFLOPs, surpassing both general and task specialized segmentation networks. Cross dataset tests confirm its strong generalization across varied scenes. The lightweight and accurate architecture provides a reliable visual solution for automated inspection in power grid monitoring systems.","manuscriptTitle":"Integrating Convexity and DeepLabv3+ for SemanticSegmentation of Power Lines","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-21 20:41:33","doi":"10.21203/rs.3.rs-8607246/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-02-27T10:52:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"338417574708145305898859561305462557531","date":"2026-02-25T05:08:17+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-20T04:02:29+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-19T23:23:27+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-16T10:00:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-16T09:58:23+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-15T05:48:15+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":"6244bf2d-9bfa-4d6b-818d-d9a944323432","owner":[],"postedDate":"January 21st, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":61441961,"name":"Physical sciences/Engineering"},{"id":61441962,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-01-21T20:41:33+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-21 20:41:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8607246","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8607246","identity":"rs-8607246","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}