MML: MoE-Based Fusion of Mamba and LightTS for Sequence Prediction in Hyperspectral Image Compression

MML: MoE-Based Fusion of Mamba and LightTS for Sequence Prediction in Hyperspectral Image Compression | 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 MML: MoE-Based Fusion of Mamba and LightTS for Sequence Prediction in Hyperspectral Image Compression Lei Zhang, Lixin Zhao, Jiaqi Wang, Wenlong Xia This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7909806/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 Hyperspectral imagery provides rich spectral–spatial information crucial for Earth observation, but its massive data volume poses severe challenges for onboard transmission and storage. The widely used CCSDS-123.0-B-2 standard (low complexity, engineering feasibility) suffers from adaptive filtering predictor limitations: poor modeling of intricate spatio-spectral correlations and slow convergence. To address these issues, this paper introduces a novel predictor architecture, termed MML (Mixture-of-Experts fused Mamba– LightTS predictor), grounded in a Mixture-of-Experts (MoE) framework. The proposed design integrates the long-sequence modeling capability of the Mamba state-space model with the lightweight efficiency of LightTS, while employing a dynamic routing mechanism to adaptively determine expert contributions. Unlike autoencoder-based schemes, it directly performs pixel-level prediction and entropy coding for lossless and near-lossless compression, thereby enhancing prediction accuracy and resource efficiency while keeping computational complexity low. Experimental evaluations on NASA AVIRIS and China’s Gaofen-5 (GF-5) datasets demonstrate that the proposed method consistently outperforms CCSDS-123.0-B-2 in terms of compression ratio and error control under both lossless and high-fidelity scenarios. Furthermore, in complex scenes, it exhibits superior performance and robustness compared with state-of-the-art deep learning approaches, including Verdu, 1D-CAE, and SSCNet. These findings provide a promising avenue for onboard hyperspectral image compression, striking a balance between efficiency, low complexity, and support for lossless operation. hyperspectral image compression CCSDS123.0-B-2 Mixture-of-Experts Mamba LightTS 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-7909806","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":542276109,"identity":"f05a19ac-88c7-4cde-afb3-d351c1bc5102","order_by":0,"name":"Lei Zhang","email":"","orcid":"","institution":"Shenyang Aerospace University","correspondingAuthor":false,"prefix":"","firstName":"Lei","middleName":"","lastName":"Zhang","suffix":""},{"id":542276112,"identity":"ce943aa5-9de2-4353-ae9e-af2089e54470","order_by":1,"name":"Lixin Zhao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+ElEQVRIiWNgGAWjYDACCSB+YMDAw8DegCqIX0sCSAvPAcYGZEECWsCMBCK1yM9ufvggoeCwjLnk8+ePeRjq5A0OMB+8zcNgl4dLC+OcY8YGCQZpPJazcwybeRgOG244wJZszcOQXIxLC7NEgplEgoENj8HtHEaglgOMGw7wmEkDGYkNOLSwSaR/A2qR4DG4efwhUEud/YYD/N/wauGRyIHacoMB5DDmRKAtbHi1SEjkFIP9YnAmx3DmHIPDyTMPsxlbzjFIxqlFfkb6xgcf/hy2Nzh+/MGHNxV1tn3Hmx/eeFNhh1MLGjAAhQiMMQpGwSgYBaOAbAAAsKFPWkjn7BMAAAAASUVORK5CYII=","orcid":"","institution":"Shenyang Aerospace University","correspondingAuthor":true,"prefix":"","firstName":"Lixin","middleName":"","lastName":"Zhao","suffix":""},{"id":542276113,"identity":"e3fa219c-bcec-4450-a502-a4655f1f27e1","order_by":2,"name":"Jiaqi Wang","email":"","orcid":"","institution":"Shenyang Aerospace University","correspondingAuthor":false,"prefix":"","firstName":"Jiaqi","middleName":"","lastName":"Wang","suffix":""},{"id":542276115,"identity":"a0f1db01-47df-474c-8f24-8d980ee57cd1","order_by":3,"name":"Wenlong Xia","email":"","orcid":"","institution":"Shenyang Aerospace University","correspondingAuthor":false,"prefix":"","firstName":"Wenlong","middleName":"","lastName":"Xia","suffix":""}],"badges":[],"createdAt":"2025-10-21 02:53:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7909806/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7909806/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95668563,"identity":"5620fada-23e1-4023-997e-1c348f52824b","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5760,"visible":true,"origin":"","legend":"","description":"","filename":"281a3dc651474e9cbf7055c1653b3e5f.json","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/b840491fa3ab1f2aecf74a09.json"},{"id":95668556,"identity":"7f6fb51d-b08f-473c-8855-6affa6d4d98d","added_by":"auto","created_at":"2025-11-11 17:06:31","extension":"png","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124042,"visible":true,"origin":"","legend":"","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/0460917ede664c2376497b08.png"},{"id":95668558,"identity":"cf0c1a11-4c35-47b7-a272-d8b47099adbb","added_by":"auto","created_at":"2025-11-11 17:06:31","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":680287,"visible":true,"origin":"","legend":"","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/d538132294a974a9f338d399.png"},{"id":95668573,"identity":"ebed7cc9-e91d-4886-8945-d761846d939e","added_by":"auto","created_at":"2025-11-11 17:06:33","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":114694,"visible":true,"origin":"","legend":"","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/f948da9ff47f9c1fd7a0f625.png"},{"id":95798048,"identity":"f72098bf-dd8a-46f9-b0a6-f6e86515c776","added_by":"auto","created_at":"2025-11-13 08:14:53","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":93927,"visible":true,"origin":"","legend":"","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/981df3f3bf8fa8672d38cffd.png"},{"id":95668559,"identity":"254bde5c-af11-4e80-b28a-18203ab3129e","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":437385,"visible":true,"origin":"","legend":"","description":"","filename":"Fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/25afdb8158be21d07bae3481.png"},{"id":95668567,"identity":"eb553ea9-a293-46fb-8b30-c4a0975b79da","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2516134,"visible":true,"origin":"","legend":"","description":"","filename":"Fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/aeaf79d142a72bd8ae4931bb.png"},{"id":95799028,"identity":"4a0068bd-0f36-4405-be4c-caf2aafbf7fd","added_by":"auto","created_at":"2025-11-13 08:18:26","extension":"pdf","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4174722,"visible":true,"origin":"","legend":"","description":"","filename":"MMLMoEBasedFusionofMambaandLightTSforSequencePredictioninHyperspectralImageCompression.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/ae14fda9f957028a7592c131.pdf"},{"id":95668566,"identity":"b1799b9b-ec6f-4448-b510-389162603fbb","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4174722,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/84cc87bd87069f1b1d24d00d.pdf"},{"id":95668560,"identity":"013cca09-81b4-4508-ba45-91f7da08fd29","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"eps","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":286,"visible":true,"origin":"","legend":"","description":"","filename":"example.eps","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/ef6cd8e4f755ad4973f0212c.eps"},{"id":95797806,"identity":"37014104-e333-4e10-b80b-397482f329c4","added_by":"auto","created_at":"2025-11-13 08:11:06","extension":"txt","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1380,"visible":true,"origin":"","legend":"","description":"","filename":"history.txt","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/af0282c74d28b18657741888.txt"},{"id":95668564,"identity":"399c8db2-2957-42fa-b497-c78439b9d3da","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"txt","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1707,"visible":true,"origin":"","legend":"","description":"","filename":"readme.txt","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/2f9d6544a4ad377fb658cbd6.txt"},{"id":95668565,"identity":"42b36586-6eb4-4b47-8b5e-30b059c1159c","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"bst","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":33238,"visible":true,"origin":"","legend":"","description":"","filename":"spbasic.bst","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/863d65b8ee3f2b37271604ea.bst"},{"id":95668561,"identity":"928f15b0-2124-41b2-a370-d3c3b24b709e","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"bst","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30130,"visible":true,"origin":"","legend":"","description":"","filename":"spmpsci.bst","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/ae9604403fefa75d7b316fdb.bst"},{"id":95797661,"identity":"40561ae2-e5c2-433b-a4f6-d1156e38518d","added_by":"auto","created_at":"2025-11-13 08:08:38","extension":"bst","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28600,"visible":true,"origin":"","legend":"","description":"","filename":"spphys.bst","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/2ea49df787b6cebab2f349f2.bst"},{"id":95668570,"identity":"1c27b964-f05f-406d-884c-bfa92f2d3048","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"clo","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3696,"visible":true,"origin":"","legend":"","description":"","filename":"svglov3.clo","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/5375f44b116dd735b8b9e37a.clo"},{"id":95797159,"identity":"0aa77522-e417-4755-b20c-1212bcc23e0c","added_by":"auto","created_at":"2025-11-13 08:01:27","extension":"cls","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47679,"visible":true,"origin":"","legend":"","description":"","filename":"svjour3.cls","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/be27eaf500f92607e543421e.cls"},{"id":95668569,"identity":"6f693fe7-1f12-4fe2-a2e8-3fc8b465712c","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"pdf","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":190374,"visible":true,"origin":"","legend":"","description":"","filename":"usrguid3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/6211f11fda4b811123fff5d3.pdf"},{"id":95668571,"identity":"c6ba7555-f336-475b-918a-cc9854204afb","added_by":"auto","created_at":"2025-11-11 17:06:32","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48660,"visible":true,"origin":"","legend":"","description":"","filename":"281a3dc651474e9cbf7055c1653b3e5f1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1/f30c03e12dac2e5953ad1836.xml"},{"id":108158892,"identity":"51412dd6-0646-442c-a1e6-d94d2bc36339","added_by":"auto","created_at":"2026-04-30 03:25:28","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1247097,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7909806/v1_covered_2b63282f-b5c6-48d5-b7cf-b1551a3ba87f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"MML: MoE-Based Fusion of Mamba and LightTS for Sequence Prediction in Hyperspectral Image Compression","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 compression, CCSDS123.0-B-2, Mixture-of-Experts, Mamba, LightTS","lastPublishedDoi":"10.21203/rs.3.rs-7909806/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7909806/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Hyperspectral imagery provides rich spectral–spatial information crucial for Earth observation, but its massive data volume poses severe challenges for onboard transmission and storage. The widely used CCSDS-123.0-B-2 standard (low complexity, engineering feasibility) suffers from adaptive filtering predictor limitations: poor modeling of intricate spatio-spectral correlations and slow convergence. To address these issues, this paper introduces a novel predictor architecture, termed MML (Mixture-of-Experts fused Mamba–\nLightTS predictor), grounded in a Mixture-of-Experts (MoE) framework. The proposed design integrates the long-sequence modeling capability of the Mamba state-space model with the lightweight efficiency of LightTS, while employing a dynamic routing mechanism to adaptively determine expert contributions. Unlike autoencoder-based schemes, it directly performs pixel-level prediction and entropy coding for lossless and near-lossless compression, thereby enhancing prediction accuracy and resource efficiency while keeping computational complexity low. Experimental evaluations on NASA AVIRIS and China’s Gaofen-5 (GF-5) datasets demonstrate that the proposed method consistently outperforms CCSDS-123.0-B-2 in terms of compression ratio and error control under both lossless and high-fidelity scenarios. Furthermore, in complex scenes, it exhibits superior performance and robustness compared with state-of-the-art deep learning approaches, including Verdu, 1D-CAE, and SSCNet. These findings provide a promising avenue for onboard hyperspectral image compression, striking a balance between efficiency, low complexity, and support for lossless operation.","manuscriptTitle":"MML: MoE-Based Fusion of Mamba and LightTS for Sequence Prediction in Hyperspectral Image Compression","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-11 17:06:27","doi":"10.21203/rs.3.rs-7909806/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":"5bf250a4-6e11-4d8c-9460-055317fa1aa0","owner":[],"postedDate":"November 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-30T03:25:13+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-11 17:06:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7909806","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7909806","identity":"rs-7909806","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}