Boosting Adversarial Transferability for Hyperspectral Image Classification Using Block Transformation and Weighted Intermediate Feature Divergence

Boosting Adversarial Transferability for Hyperspectral Image Classification Using Block Transformation and Weighted Intermediate Feature Divergence | 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 Boosting Adversarial Transferability for Hyperspectral Image Classification Using Block Transformation and Weighted Intermediate Feature Divergence Chun Liu, Bingqian Zhu, Tao Xu, Zheng Zheng, Zheng Li, Wei Yang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8787976/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Deep Neural Networks (DNNs) are vulnerable to adversarial attacks, which pose security challenges to hyperspectral image (HSI) classification based on DNNs. Studying adversarial attacks on DNN models for HSI classification helps reveal model vulnerabilities but also provides critical directions for improving their resilience. Numerous adversarial attack methods have been designed in the domain of natural images. However, different from natural images, HSIs contains high-dimensional rich spectral information, which presents new challenges for generating adversarial examples. Based on the specific characteristics of HSIs, this paper proposes a novel method to enhance the transferability of the adversarial examples for HSI classification using block transformation and weighted intermediate feature divergence. While keeping the spatial structure unchanged and maintaining the global semantics, the proposed method divides the image into blocks in both spatial and spectral dimensions. Then, various transformations are applied on each block to increase input diversity and mitigate the overfitting to substitute models. Moreover, a weighted intermediate feature divergence loss is also designed by leveraging the differences between the intermediate features of original and adversarial examples, which enhances the transferability of generated adversarial examples from the perspective of optimizing the perturbation generation process. Extensive experiments demonstrate that the adversarial examples generated by the proposed method achieve more effective adversarial transferability on three public HSI datasets. Furthermore, the method maintains robust attack performance even under defense strategies. Code is available at https://github.com/Moon-star1/Boosting-Adversarial-Transferability-for-HSI-Classification. Hyperspectral image classification Adversarial examples Adversarial transferability Block transformation Feature divergence Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted 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-8787976","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":610678689,"identity":"604629d1-f029-412d-9069-c7dcc5c475b2","order_by":0,"name":"Chun Liu","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Chun","middleName":"","lastName":"Liu","suffix":""},{"id":610678690,"identity":"7efb3a56-d94c-43e1-a233-fd3096937845","order_by":1,"name":"Bingqian Zhu","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Bingqian","middleName":"","lastName":"Zhu","suffix":""},{"id":610678691,"identity":"16ff1310-b5f7-419a-bad2-6aee1d9d7178","order_by":2,"name":"Tao Xu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAqUlEQVRIiWNgGAWjYHACxsdQhgFx6nkYGJiNSdbCJk2aFnuJHLPqgoo7iQ3szdskGGruEGELzxmz2zPOPEts4DlWJsFw7BkRWth7zG7zth1ObABaJ8HYcJgILcw8ZsW8/4Ba5N8QqwVoCzNvA8gWHmK1nDlWLM1z7LBxG09asUXCMSK0sM9I3viZp+awbD/74Y03PtQQoQUO2EBEAgkaRsEoGAWjYBTgAQCv3jKq+sP0jwAAAABJRU5ErkJggg==","orcid":"","institution":"Henan University","correspondingAuthor":true,"prefix":"","firstName":"Tao","middleName":"","lastName":"Xu","suffix":""},{"id":610678692,"identity":"8cef250c-6e8d-4438-8b4d-7da70e2e2cd9","order_by":3,"name":"Zheng Zheng","email":"","orcid":"","institution":"Beihang University","correspondingAuthor":false,"prefix":"","firstName":"Zheng","middleName":"","lastName":"Zheng","suffix":""},{"id":610678693,"identity":"6e816f1f-a295-4ef7-9c79-ab984a08c40f","order_by":4,"name":"Zheng Li","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Zheng","middleName":"","lastName":"Li","suffix":""},{"id":610678694,"identity":"cb2257e4-e9c5-4209-bca4-bab2516c5de3","order_by":5,"name":"Wei Yang","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Yang","suffix":""},{"id":610678695,"identity":"761d4a13-7737-4cea-98d3-36b793c3bb18","order_by":6,"name":"Zhigang Han","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Zhigang","middleName":"","lastName":"Han","suffix":""},{"id":610678696,"identity":"8aba3a3a-12c0-4319-9774-9896e8e2b0b9","order_by":7,"name":"Jiayao Wang","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Jiayao","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2026-02-04 14:55:47","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8787976/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8787976/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106503353,"identity":"b4430c31-5ffe-4f17-9804-06b091c95eb0","added_by":"auto","created_at":"2026-04-09 09:29:36","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6406169,"visible":true,"origin":"","legend":"","description":"","filename":"Knowledge.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8787976/v1_covered_874ec2f1-51d2-4f65-b55b-113157587f22.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Boosting Adversarial Transferability for Hyperspectral Image Classification Using Block Transformation and Weighted Intermediate Feature Divergence","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"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":"Hyperspectral image classification, Adversarial examples, Adversarial transferability, Block transformation, Feature divergence","lastPublishedDoi":"10.21203/rs.3.rs-8787976/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8787976/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDeep Neural Networks (DNNs) are vulnerable to adversarial attacks, which pose security challenges to hyperspectral image (HSI) classification based on DNNs. Studying adversarial attacks on DNN models for HSI classification helps reveal model vulnerabilities but also provides critical directions for improving their resilience. Numerous adversarial attack methods have been designed in the domain of natural images. However, different from natural images, HSIs contains high-dimensional rich spectral information, which presents new challenges for generating adversarial examples. Based on the specific characteristics of HSIs, this paper proposes a novel method to enhance the transferability of the adversarial examples for HSI classification using block transformation and weighted intermediate feature divergence. While keeping the spatial structure unchanged and maintaining the global semantics, the proposed method divides the image into blocks in both spatial and spectral dimensions. Then, various transformations are applied on each block to increase input diversity and mitigate the overfitting to substitute models. Moreover, a weighted intermediate feature divergence loss is also designed by leveraging the differences between the intermediate features of original and adversarial examples, which enhances the transferability of generated adversarial examples from the perspective of optimizing the perturbation generation process. Extensive experiments demonstrate that the adversarial examples generated by the proposed method achieve more effective adversarial transferability on three public HSI datasets. Furthermore, the method maintains robust attack performance even under defense strategies. Code is available at https://github.com/Moon-star1/Boosting-Adversarial-Transferability-for-HSI-Classification.\u003c/p\u003e","manuscriptTitle":"Boosting Adversarial Transferability for Hyperspectral Image Classification Using Block Transformation and Weighted Intermediate Feature Divergence","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-25 05:52:42","doi":"10.21203/rs.3.rs-8787976/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":"46aa649b-fdd3-4788-a5e1-ea373892a81d","owner":[],"postedDate":"March 25th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-09T09:28:08+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-25 05:52:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8787976","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8787976","identity":"rs-8787976","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}