Dynamic-Static Hybrid Dictionary Learning: Enhancing Deep K-SVD for Image Denoising and Beyond

Dynamic-Static Hybrid Dictionary Learning: Enhancing Deep K-SVD for Image Denoising and Beyond | 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 Dynamic-Static Hybrid Dictionary Learning: Enhancing Deep K-SVD for Image Denoising and Beyond Zhonggui Sun, Jing Yao, Can Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7701964/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 23 Jan, 2026 Read the published version in The Visual Computer → Version 1 posted 8 You are reading this latest preprint version Abstract Dictionary learning (DicL) is a fundamental technique in sparse representation, widely applied in image processing. As a promising deep extension of traditional DicL, Deep K-SVD (DKSVD) inherits the interpretability of classical models while benefiting from the strong learning capacity of deep networks. However, its reliance on static dictionaries limits adaptability in complex scenarios. To overcome this limitation, we propose DS-DKSVD, a dynamic-static extension of DKSVD, which integrates a hybrid dictionary composed of static and dynamic components. The static component, represented by network parameters, captures global features from training data, while the dynamic component, generated by a dedicated sub-network, adapts to specific input characteristics. During patch averaging, DS-DKSVD dynamically assigns weights, enhancing inter-patch variation handling. Extensive experiments on non-blind and blind image denoising demonstrate its superiority over existing methods. DS-DKSVD achieves up to 0.46 dB and 0.42 dB improvements in PSNR over the original DKSVD and its adaptive variant (AKSVD), respectively. Beyond denoising, a preliminary image classification task highlights the broader applicability of DS-DKSVD. Complementing these quantitative results, visualizations of the learned hybrid dictionary provide qualitative evidence of its interpretability, revealing the complementary roles of static and dynamic components. The source code for the DS-DKSVD is publicly available at https://github.com/yaojingzeo/DS-DKSVD . Deep K-SVD deep dictionary learning dynamic strategies sparse representations image denoising Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 23 Jan, 2026 Read the published version in The Visual Computer → Version 1 posted Editorial decision: Revision requested 28 Oct, 2025 Reviewers agreed at journal 21 Oct, 2025 Reviews received at journal 20 Oct, 2025 Reviewers agreed at journal 30 Sep, 2025 Reviewers invited by journal 30 Sep, 2025 Editor assigned by journal 25 Sep, 2025 Submission checks completed at journal 24 Sep, 2025 First submitted to journal 24 Sep, 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-7701964","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":519875033,"identity":"65d9bd4d-f9bd-4fdc-be73-0b6a8e5d6d06","order_by":0,"name":"Zhonggui Sun","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwElEQVRIiWNgGAWjYDACdgbGAwwVB8DsAw+I0sLMzHCA4cwBBh6QlgSitTC2QbQwEKWFv5n/wOHCeXfk7MUOPwTaYien20BAi8RhZobDM7c9M+aRTjMAakk2NjtAyBqQFt5thxN7pBNAWg4kbiOkRR6sZQ5IS/oH4rQYgLU0gLTkEGmL4WFmg8Mzjh025rmdU3AgwYAIv8gdb3z4uKDmsBz77PTNHz5U2MkR9j4QMCO5kwjlaFpGwSgYBaNgFGABAGQ3RrUHHs7GAAAAAElFTkSuQmCC","orcid":"","institution":"Liaocheng University","correspondingAuthor":true,"prefix":"","firstName":"Zhonggui","middleName":"","lastName":"Sun","suffix":""},{"id":519875034,"identity":"56b12c20-4685-4aa4-b528-c649530c4384","order_by":1,"name":"Jing Yao","email":"","orcid":"","institution":"Liaocheng University","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Yao","suffix":""},{"id":519875035,"identity":"21135648-c32f-4291-b26f-6bfad999d44a","order_by":2,"name":"Can Zhang","email":"","orcid":"","institution":"Chongqing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Can","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-09-24 09:38:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7701964/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7701964/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00371-026-04361-w","type":"published","date":"2026-01-23T15:57:45+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":92160428,"identity":"4389a7a0-77a2-444f-8b81-0a32621939c1","added_by":"auto","created_at":"2025-09-25 09:54:13","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5374,"visible":true,"origin":"","legend":"","description":"","filename":"a88b48eb64844d2eb0fa6beaaa97f6b3.json","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/95d20fad08cb0e7dc9cbb325.json"},{"id":92160429,"identity":"fb643bda-d56a-4d86-b69d-e0fac264fc10","added_by":"auto","created_at":"2025-09-25 09:54:14","extension":"xml","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":169556,"visible":true,"origin":"","legend":"","description":"","filename":"a88b48eb64844d2eb0fa6beaaa97f6b31enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/21844c0aac6877f2e8f867f1.xml"},{"id":92160431,"identity":"e0e7616d-e35d-4104-b420-b0bc5f332906","added_by":"auto","created_at":"2025-09-25 09:54:14","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":62373,"visible":true,"origin":"","legend":"","description":"","filename":"CanZhang.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/10312731a3f7df0d3f3f687b.jpg"},{"id":92160443,"identity":"1da43f61-5480-4643-9730-d06def29da01","added_by":"auto","created_at":"2025-09-25 09:54:15","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":86088756,"visible":true,"origin":"","legend":"","description":"","filename":"DSDKSVD.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/54aa7834722c74f20979ca98.pdf"},{"id":92160788,"identity":"655ee0ad-9a08-400a-bb43-eb78b36c5d6b","added_by":"auto","created_at":"2025-09-25 10:02:14","extension":"jpg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":106608,"visible":true,"origin":"","legend":"","description":"","filename":"JingYao.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/d47782c7da1fe743d814d87a.jpg"},{"id":92160438,"identity":"0b458a0d-ccd1-4c6b-bcfb-bfca75e766fc","added_by":"auto","created_at":"2025-09-25 09:54:14","extension":"jpg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49083,"visible":true,"origin":"","legend":"","description":"","filename":"ZhongguiSun.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/9285698b58ba6993e2346767.jpg"},{"id":92160436,"identity":"5899b4d9-6e42-482e-b7d5-b64bf34fda74","added_by":"auto","created_at":"2025-09-25 09:54:14","extension":"cls","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":55857,"visible":true,"origin":"","legend":"","description":"","filename":"snjnl.cls","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/ca9f981c69eb921bb13511fa.cls"},{"id":92160432,"identity":"aee67520-a206-4312-98c2-30fbb238468d","added_by":"auto","created_at":"2025-09-25 09:54:14","extension":"bst","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64166,"visible":true,"origin":"","legend":"","description":"","filename":"snmathphysnum.bst","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/6a29af6eb795433ce8988de4.bst"},{"id":92160433,"identity":"18d85e08-0afa-4b0a-9d8c-0a118644c6dd","added_by":"auto","created_at":"2025-09-25 09:54:14","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":90467,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineCanZhang.png","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/59f4bfbac20e32c95693fed1.png"},{"id":92160434,"identity":"f50d49ff-4eba-4ba4-bbff-ed72feb24e3a","added_by":"auto","created_at":"2025-09-25 09:54:14","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":760902,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineJingYao.png","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/2c3e622f878da240cd2ff038.png"},{"id":92160439,"identity":"ec315de0-621e-40d9-89f4-ac77a93959c7","added_by":"auto","created_at":"2025-09-25 09:54:14","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":255078,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineZhongguiSun.png","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/aa6147595f7132b436aeb11c.png"},{"id":92160789,"identity":"d607d02f-7901-4bf9-b6f3-760deeb6c1e2","added_by":"auto","created_at":"2025-09-25 10:02:14","extension":"xml","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":178276,"visible":true,"origin":"","legend":"","description":"","filename":"a88b48eb64844d2eb0fa6beaaa97f6b31structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/e5a8890777c3210036238a34.xml"},{"id":92160790,"identity":"81719f9e-0143-4794-a922-a36c1bf85813","added_by":"auto","created_at":"2025-09-25 10:02:14","extension":"html","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":205033,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1/f5466276283a8a65ccd9cd49.html"},{"id":101151899,"identity":"baf80d30-4f50-40ed-b5e1-1f13bc46592e","added_by":"auto","created_at":"2026-01-26 16:07:33","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4792994,"visible":true,"origin":"","legend":"","description":"","filename":"DSDKSVD.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7701964/v1_covered_8ffabec6-f9a0-4676-9dd5-0b8f6c7bdde7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Dynamic-Static Hybrid Dictionary Learning: Enhancing Deep K-SVD for Image Denoising and Beyond","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"the-visual-computer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"tvcj","sideBox":"Learn more about [The Visual Computer](http://link.springer.com/journal/371)","snPcode":"371","submissionUrl":"https://submission.nature.com/new-submission/371/3","title":"The Visual Computer","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Deep K-SVD, deep dictionary learning, dynamic strategies, sparse representations, image denoising","lastPublishedDoi":"10.21203/rs.3.rs-7701964/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7701964/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDictionary learning (DicL) is a fundamental technique in sparse representation, widely applied in image processing. As a promising deep extension of traditional DicL, Deep K-SVD (DKSVD) inherits the interpretability of classical models while benefiting from the strong learning capacity of deep networks. However, its reliance on static dictionaries limits adaptability in complex scenarios. To overcome this limitation, we propose DS-DKSVD, a dynamic-static extension of DKSVD, which integrates a hybrid dictionary composed of static and dynamic components. The static component, represented by network parameters, captures global features from training data, while the dynamic component, generated by a dedicated sub-network, adapts to specific input characteristics. During patch averaging, DS-DKSVD dynamically assigns weights, enhancing inter-patch variation handling. Extensive experiments on non-blind and blind image denoising demonstrate its superiority over existing methods. DS-DKSVD achieves up to 0.46 dB and 0.42 dB improvements in PSNR over the original DKSVD and its adaptive variant (AKSVD), respectively. Beyond denoising, a preliminary image classification task highlights the broader applicability of DS-DKSVD. Complementing these quantitative results, visualizations of the learned hybrid dictionary provide qualitative evidence of its interpretability, revealing the complementary roles of static and dynamic components. The source code for the DS-DKSVD is publicly available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/yaojingzeo/DS-DKSVD\u003c/span\u003e\u003cspan address=\"https://github.com/yaojingzeo/DS-DKSVD\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e","manuscriptTitle":"Dynamic-Static Hybrid Dictionary Learning: Enhancing Deep K-SVD for Image Denoising and Beyond","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-25 09:54:09","doi":"10.21203/rs.3.rs-7701964/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-28T22:02:00+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"87402368518420097157897230030882255859","date":"2025-10-21T11:08:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-20T05:50:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"74155588268627584653903860268561708862","date":"2025-10-01T03:08:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-30T20:08:50+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-25T11:28:05+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-25T03:16:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"The Visual Computer","date":"2025-09-24T09:23:14+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"the-visual-computer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"tvcj","sideBox":"Learn more about [The Visual Computer](http://link.springer.com/journal/371)","snPcode":"371","submissionUrl":"https://submission.nature.com/new-submission/371/3","title":"The Visual Computer","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d91d0523-6daf-4a09-8d7b-fe3faeb2b018","owner":[],"postedDate":"September 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-01-26T16:03:40+00:00","versionOfRecord":{"articleIdentity":"rs-7701964","link":"https://doi.org/10.1007/s00371-026-04361-w","journal":{"identity":"the-visual-computer","isVorOnly":false,"title":"The Visual Computer"},"publishedOn":"2026-01-23 15:57:45","publishedOnDateReadable":"January 23rd, 2026"},"versionCreatedAt":"2025-09-25 09:54:09","video":"","vorDoi":"10.1007/s00371-026-04361-w","vorDoiUrl":"https://doi.org/10.1007/s00371-026-04361-w","workflowStages":[]},"version":"v1","identity":"rs-7701964","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7701964","identity":"rs-7701964","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}