Post-Disaster Affected Area Segmentation with Vision Transformer (ViT)-based Model using Sentinel-2 and Formosat-5 Imagery

Post-Disaster Affected Area Segmentation with Vision Transformer (ViT)-based Model using Sentinel-2 and Formosat-5 Imagery | 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 Post-Disaster Affected Area Segmentation with Vision Transformer (ViT)-based Model using Sentinel-2 and Formosat-5 Imagery Yi-Shan Chu, Hsuan-Cheng Wei This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7214554/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract We propose a vision transformer (ViT)-based deep learning framework to improve disaster-affected area segmentation from satellite images, supporting the Emergent Value Added Product (EVAP) system developed by the Taiwan Space Agency (TASA). The process begins with a small number of manually labeled regions. We then use principal component analysis (PCA) to expand these labels with a confidence interval, creating a weakly supervised training set. Our model, which takes multi-band input from Sentinel-2 and Formosat-5 satellites, is trained to distinguish disaster-affected areas using these expanded labels. We adopt several strategies to increase accuracy when only limited supervision is available. To evaluate performance, our predictions are compared to higher-resolution EVAP results to measure spatial accuracy and consistency. Experiments on real disaster events, such as the 2022 Poyang Lake drought and the 2023 Rhodes wildfire, shows that our approach produces smoother and more reliable segmentation maps, providing a practical solution for disaster mapping when detailed ground truth is lacking. Remote sensing imagery Post disaster analysis Change detection Vision Transformer (ViT) Sentinel-2 Formosat-5 Principal Component Analysis (PCA) Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 05 Oct, 2025 Reviews received at journal 30 Aug, 2025 Reviews received at journal 28 Aug, 2025 Reviewers agreed at journal 06 Aug, 2025 Reviewers agreed at journal 05 Aug, 2025 Reviewers invited by journal 05 Aug, 2025 Editor assigned by journal 28 Jul, 2025 Submission checks completed at journal 28 Jul, 2025 First submitted to journal 25 Jul, 2025 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-7214554","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":496460569,"identity":"564f0727-183f-43f6-bc86-8fa8dc66b10e","order_by":0,"name":"Yi-Shan Chu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+0lEQVRIiWNgGAWjYBACNmbG9h8JBhJ29scb4IIGeLXwsTM3SHwosElmOHMAWUsCbi1y/OwNkjM+pDE23ECowq8F6LAGYx6Dw8yMM59fYC6osUtsYG/eJsH44zBeLclALXzM0jkFzDOOJSc28Bwrk2BIwK/lMMgWNumcBGbehgOJDRI5ZkAtt/FpaWwGamHskTwD1SL/hqCWZsYZBmmMMyTYD0Bt4SGopY3hg4FNsgFPDsNhnmPJxm08acUWCWn/cWqR7z/+jCHhj4SdAfvxh495auxk+9kPb7zxwSYNpxYkwGNwAGwviEggRgMDA/sD4tSNglEwCkbBiAMAjqBNbPuOGzwAAAAASUVORK5CYII=","orcid":"","institution":"National Chengchi University","correspondingAuthor":true,"prefix":"","firstName":"Yi-Shan","middleName":"","lastName":"Chu","suffix":""},{"id":496460570,"identity":"da0a47dc-1e33-45d3-9444-75d56b7419b5","order_by":1,"name":"Hsuan-Cheng Wei","email":"","orcid":"","institution":"Taiwan Space Agency","correspondingAuthor":false,"prefix":"","firstName":"Hsuan-Cheng","middleName":"","lastName":"Wei","suffix":""}],"badges":[],"createdAt":"2025-07-25 13:08:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7214554/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7214554/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88629165,"identity":"9e3d0433-4242-49b8-96bc-e1dc94aba68f","added_by":"auto","created_at":"2025-08-08 13:28:06","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5234938,"visible":true,"origin":"","legend":"","description":"","filename":"JournalofRemoteSensing.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7214554/v1_covered_82ddfc30-b77d-4676-9d1b-11a03ae3a5d7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Post-Disaster Affected Area Segmentation with Vision Transformer (ViT)-based Model using Sentinel-2 and Formosat-5 Imagery","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":"terrestrial-atmospheric-and-oceanic-sciences","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"taoj","sideBox":"Learn more about [Terrestrial, Atmospheric and Oceanic Sciences](https://link.springer.com/journal/44195)","snPcode":"44195","submissionUrl":"https://submission.springernature.com/new-submission/44195/3","title":"Terrestrial, Atmospheric and Oceanic Sciences","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Open","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Remote sensing imagery, Post disaster analysis, Change detection, Vision Transformer (ViT), Sentinel-2, Formosat-5, Principal Component Analysis (PCA)","lastPublishedDoi":"10.21203/rs.3.rs-7214554/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7214554/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"We propose a vision transformer (ViT)-based deep learning framework to improve disaster-affected area segmentation from satellite images, supporting the Emergent Value Added Product (EVAP) system developed by the Taiwan Space Agency (TASA). The process begins with a small number of manually labeled regions. We then use principal component analysis (PCA) to expand these labels with a confidence interval, creating a weakly supervised training set. Our model, which takes multi-band input from Sentinel-2 and Formosat-5 satellites, is trained to distinguish disaster-affected areas using these expanded labels. We adopt several strategies to increase accuracy when only limited supervision is available. To evaluate performance, our predictions are compared to higher-resolution EVAP results to measure spatial accuracy and consistency. Experiments on real disaster events, such as the 2022 Poyang Lake drought and the 2023 Rhodes wildfire, shows that our approach produces smoother and more reliable segmentation maps, providing a practical solution for disaster mapping when detailed ground truth is lacking.","manuscriptTitle":"Post-Disaster Affected Area Segmentation with Vision Transformer (ViT)-based Model using Sentinel-2 and Formosat-5 Imagery","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-08 13:20:01","doi":"10.21203/rs.3.rs-7214554/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-05T11:43:40+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-30T16:23:02+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-28T05:34:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"231248635551176482979681283471808547482","date":"2025-08-06T08:23:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"212258454565054962333414100749937260856","date":"2025-08-05T07:45:06+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-05T06:22:11+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-28T13:46:23+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-28T13:45:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Terrestrial, Atmospheric and Oceanic Sciences","date":"2025-07-25T12:52:19+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"terrestrial-atmospheric-and-oceanic-sciences","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"taoj","sideBox":"Learn more about [Terrestrial, Atmospheric and Oceanic Sciences](https://link.springer.com/journal/44195)","snPcode":"44195","submissionUrl":"https://submission.springernature.com/new-submission/44195/3","title":"Terrestrial, Atmospheric and Oceanic Sciences","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Open","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"71c0cda8-b22c-4b2b-ace8-28bb596fcb56","owner":[],"postedDate":"August 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-01-21T23:38:16+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-08 13:20:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7214554","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7214554","identity":"rs-7214554","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}