FARR: An Efficient Frozen-Feature Learning Framework for Wood Species Identification with Applications to Texture Recognition

FARR: An Efficient Frozen-Feature Learning Framework for Wood Species Identification with Applications to Texture Recognition | 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 FARR: An Efficient Frozen-Feature Learning Framework for Wood Species Identification with Applications to Texture Recognition Tao Yang, Rigui Zhou, Pengju Ren, Hongpeng Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7972880/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Jan, 2026 Read the published version in Signal, Image and Video Processing → Version 1 posted 9 You are reading this latest preprint version Abstract The scarcity of high-value timber resources and increasing market fraud have intensified the demand for efficient wood species identification technologies. Traditional methods and deep learning approaches using X-ray imaging are constrained by efficiency limitations and insufficient training samples. This paper proposes FARR, a wood species identification framework that completely freezes pre-trained networks as feature extractors. The framework achieves efficient feature learning through cross-block feature aggregation and learnable residual connections , attention mechanisms, randomized autoencoders, with low computational complexity requiring optimization of only a few parameters. Experimental results demonstrate that the proposed method achieves 99.86% accuracy on wood identification tasks, with training efficiency improved by 20–68× over partial fine-tuning by freezing the backbone and reducing parameter updates. The method exhibits excellent robustness and generalization capability across multiple datasets and complex environmental conditions. This study provides an efficient solution for intelligent wood identification in resource-constrained scenarios, reducing computational costs while maintaining identification accuracy and offering new insights for deep learning applications under small-sample conditions. Texture recognition Feature fusion Computational efficiency Frozen features Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 16 Jan, 2026 Read the published version in Signal, Image and Video Processing → Version 1 posted Editorial decision: Revision requested 19 Nov, 2025 Reviews received at journal 17 Nov, 2025 Reviews received at journal 02 Nov, 2025 Reviewers agreed at journal 30 Oct, 2025 Reviewers agreed at journal 30 Oct, 2025 Reviewers invited by journal 30 Oct, 2025 Editor assigned by journal 29 Oct, 2025 Submission checks completed at journal 29 Oct, 2025 First submitted to journal 28 Oct, 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-7972880","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":542302290,"identity":"f928c6bd-a19e-4c15-a226-22c09729e8d0","order_by":0,"name":"Tao Yang","email":"","orcid":"","institution":"Shanghai Maritime University","correspondingAuthor":false,"prefix":"","firstName":"Tao","middleName":"","lastName":"Yang","suffix":""},{"id":542302292,"identity":"e4f14176-8504-46f4-983d-ab864fc07a3e","order_by":1,"name":"Rigui Zhou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYHACNiC2YTCAcJiJ1pIG1MJMmpbDJGgx71/+7MHHtvP25mLnj0kwVFgnNrCfPYBXi8yNB+mGM9tuJ+6cncwmwXAmPbGBJy8BrxYJiQPHpHnbbicY3AZqYWw7nNggwWNAQMvBNqCWc/YQLf+I0cLfzAbUcoBxA1hLA1G2sLFJzjiXnAjUYmyRcCzduI0nh5Atx59JfCizAzos8eGNDzXWsv3sZ/BrYZBIQOKA2Gz41QMB/wGCSkbBKBgFo2CkAwB9/0A7HRppOgAAAABJRU5ErkJggg==","orcid":"","institution":"Shanghai Maritime University","correspondingAuthor":true,"prefix":"","firstName":"Rigui","middleName":"","lastName":"Zhou","suffix":""},{"id":542302293,"identity":"f5fd0cfc-7b4c-4b73-83e2-f2bb2180af71","order_by":2,"name":"Pengju Ren","email":"","orcid":"","institution":"Shanghai Maritime University","correspondingAuthor":false,"prefix":"","firstName":"Pengju","middleName":"","lastName":"Ren","suffix":""},{"id":542302295,"identity":"641a28f6-550f-4527-9a33-82b1021cd2e1","order_by":3,"name":"Hongpeng Wang","email":"","orcid":"","institution":"Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Hongpeng","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-10-28 18:09:57","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7972880/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7972880/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11760-025-05078-4","type":"published","date":"2026-01-16T16:30:34+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":95656037,"identity":"82f53e73-d15e-42d1-86fc-af55f1600e92","added_by":"auto","created_at":"2025-11-11 16:17:36","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6345,"visible":true,"origin":"","legend":"","description":"","filename":"96d7c8b00715459085cc7c849cf39702.json","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/73750770a24070befdcf0e25.json"},{"id":95593518,"identity":"af60b951-f7f6-4a59-b1a8-51f9f28025ef","added_by":"auto","created_at":"2025-11-11 03:07:41","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":156465,"visible":true,"origin":"","legend":"","description":"","filename":"CoverLetter.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/4d475a260c944d37eb566a3a.pdf"},{"id":95656043,"identity":"5b5077aa-4586-4ec4-8de4-70faf40712c4","added_by":"auto","created_at":"2025-11-11 16:17:40","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":714175,"visible":true,"origin":"","legend":"","description":"","filename":"GraphicsAbstract.png","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/afa6b3b32bb176d86b5b3124.png"},{"id":95655322,"identity":"4205ceff-1b74-48b1-b286-20c8e3ce0c92","added_by":"auto","created_at":"2025-11-11 16:15:25","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":100849,"visible":true,"origin":"","legend":"","description":"","filename":"Performancecomparison.png","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/e2257541748ba49d77a8e0ab.png"},{"id":95593516,"identity":"022100a4-705b-45b0-ac36-1b812e80a6f6","added_by":"auto","created_at":"2025-11-11 03:07:41","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":61084,"visible":true,"origin":"","legend":"","description":"","filename":"Trainingefficiencyvs.accuracycomparison.png","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/486c1829c0235582d280c480.png"},{"id":95593519,"identity":"cc78e739-2d7a-496d-9084-ffaa6779d29c","added_by":"auto","created_at":"2025-11-11 03:07:41","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":94766,"visible":true,"origin":"","legend":"","description":"","filename":"framework.png","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/e057b8f634ce8c8abfdae701.png"},{"id":95656712,"identity":"b758adee-e03e-4520-ae2a-c428073031c8","added_by":"auto","created_at":"2025-11-11 16:19:25","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":583447,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/afbcd1d2db2c4ede98b000ca.pdf"},{"id":95593520,"identity":"a42619ed-b475-4094-9277-50d3ed2b6126","added_by":"auto","created_at":"2025-11-11 03:07:41","extension":"cls","order_by":7,"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-7972880/v1/1a6f908a9a9926bf7de5152e.cls"},{"id":95593526,"identity":"ddd6bcaf-12a0-4268-aed8-dfaacaf61760","added_by":"auto","created_at":"2025-11-11 03:07:41","extension":"bst","order_by":8,"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-7972880/v1/9a1009fe0ea53294f06df3d3.bst"},{"id":95593529,"identity":"780b2714-1b54-4db3-a498-3fc66855c16f","added_by":"auto","created_at":"2025-11-11 03:07:41","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":706649,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineGraphicsAbstract.png","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/a8ee9c4dd83c54607fa31f25.png"},{"id":95655567,"identity":"88e9788d-1356-4a68-86a9-2f2aa1956a87","added_by":"auto","created_at":"2025-11-11 16:16:29","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":62386,"visible":true,"origin":"","legend":"","description":"","filename":"OnlinePerformancecomparison.png","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/5c5f78fca3e67809b41b2af6.png"},{"id":95656158,"identity":"16863c4c-2265-4e04-96a3-e08b91e26196","added_by":"auto","created_at":"2025-11-11 16:17:53","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48069,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineTrainingefficiencyvs.accuracycomparison.png","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/f3bc39c326643b85693c354e.png"},{"id":95655576,"identity":"4ee69a3e-6db3-42e4-a8d5-2150f941b647","added_by":"auto","created_at":"2025-11-11 16:16:30","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78044,"visible":true,"origin":"","legend":"","description":"","filename":"Onlineframework.png","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/e8da0249e54245ded62c1f98.png"},{"id":95593527,"identity":"53289e5f-a6f6-45d7-8662-f3787b348f90","added_by":"auto","created_at":"2025-11-11 03:07:41","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":71576,"visible":true,"origin":"","legend":"","description":"","filename":"96d7c8b00715459085cc7c849cf397021structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1/48f24505f7b47c53ca93fcdc.xml"},{"id":100616257,"identity":"fb11d41f-31ee-404d-aeea-d8450d9c1d91","added_by":"auto","created_at":"2026-01-19 17:41:53","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":463708,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7972880/v1_covered_4ab6691a-b4aa-4aef-8c4c-6546b048202f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"FARR: An Efficient Frozen-Feature Learning Framework for Wood Species Identification with Applications to Texture Recognition","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":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Texture recognition, Feature fusion, Computational efficiency, Frozen features","lastPublishedDoi":"10.21203/rs.3.rs-7972880/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7972880/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The scarcity of high-value timber resources and increasing market fraud have intensified the demand for efficient wood species identification technologies. Traditional methods and deep learning approaches using X-ray imaging are constrained by efficiency limitations and insufficient training samples. This paper proposes FARR, a wood species identification framework that completely freezes pre-trained networks as feature extractors. The framework achieves efficient feature learning through cross-block feature aggregation and learnable residual connections , attention mechanisms, randomized autoencoders, with low computational complexity requiring optimization of only a few parameters. Experimental results demonstrate that the proposed method achieves 99.86% accuracy on wood identification tasks, with training efficiency improved by 20–68× over partial fine-tuning by freezing the backbone and reducing parameter updates. The method exhibits excellent robustness and generalization capability across multiple datasets and complex environmental conditions. This study provides an efficient solution for intelligent wood identification in resource-constrained scenarios, reducing computational costs while maintaining identification accuracy and offering new insights for deep learning applications under small-sample conditions.","manuscriptTitle":"FARR: An Efficient Frozen-Feature Learning Framework for Wood Species Identification with Applications to Texture Recognition","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-11 03:07:36","doi":"10.21203/rs.3.rs-7972880/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-19T22:08:21+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-17T05:16:13+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-02T15:14:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"233810970487964486176813843818541723964","date":"2025-10-31T02:01:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"119284292750798420817026665495607042677","date":"2025-10-31T01:30:49+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-30T23:13:54+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-29T17:38:31+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-29T17:38:22+00:00","index":"","fulltext":""},{"type":"submitted","content":"Signal, Image and Video Processing","date":"2025-10-28T17:43:11+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"7f3e1974-f07c-4b27-b3e2-e8eccb0f969e","owner":[],"postedDate":"November 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-01-19T17:08:12+00:00","versionOfRecord":{"articleIdentity":"rs-7972880","link":"https://doi.org/10.1007/s11760-025-05078-4","journal":{"identity":"signal-image-and-video-processing","isVorOnly":false,"title":"Signal, Image and Video Processing"},"publishedOn":"2026-01-16 16:30:34","publishedOnDateReadable":"January 16th, 2026"},"versionCreatedAt":"2025-11-11 03:07:36","video":"","vorDoi":"10.1007/s11760-025-05078-4","vorDoiUrl":"https://doi.org/10.1007/s11760-025-05078-4","workflowStages":[]},"version":"v1","identity":"rs-7972880","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7972880","identity":"rs-7972880","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}