InSwAV: Involution enhanced feature clustering and swapped assignments for porcelain relic microscopic image classification

InSwAV: Involution enhanced feature clustering and swapped assignments for porcelain relic microscopic image classification | 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 Article InSwAV: Involution enhanced feature clustering and swapped assignments for porcelain relic microscopic image classification Yangyang Liu, Jiatong Liu, Xinda Liu, Guohua Geng, Zhan Li This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7136122/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 24 Feb, 2026 Read the published version in npj Heritage Science → Version 1 posted 12 You are reading this latest preprint version Abstract The precise classification of porcelain relic fragments is a crucial step in the process of artifact restoration and reconstruction. The current classification methods based on apparent features such as glaze color and patterns can achieve preliminary classification, but they have obvious limitations when dealing with porcelain fragments with highly similar visual features. Especially when the sample size of microscopic images is limited, the classification performance of traditional methods often drops significantly. To address the challenges of high computational complexity and low efficiency in traditional contrastive learning for image classification, we propose InSwAV, which aligns deep features with cluster prototypes by swapped assignment through involution-enhanced feature extraction and cross-view consistency enforcement. First, we propose ResInv combining involution blocks with residual blocks of the feature extraction network. Next, features extracted by ResInv are clustered, and consistency in cluster assignments is enforced across multi-view augmented samples via cross-prediction mechanisms. This approach iteratively optimizes feature representation by minimizing cross-entropy loss, significantly reducing training time while maintaining accuracy. Furthermore, we constructed the Porcelain Relic Microscopic Images(PRMI) dataset of five classification, with data augmentation applied to enhance model robustness. Experimental results show that InSwAV achieves a classification accuracy of 96.2% on the porcelain relic microscopic image dataset, outperforming existing methods. Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 24 Feb, 2026 Read the published version in npj Heritage Science → Version 1 posted Editorial decision: Revision requested 15 Aug, 2025 Reviews received at journal 13 Aug, 2025 Reviews received at journal 13 Aug, 2025 Reviews received at journal 25 Jul, 2025 Reviewers agreed at journal 25 Jul, 2025 Reviewers agreed at journal 25 Jul, 2025 Reviewers agreed at journal 24 Jul, 2025 Reviewers agreed at journal 23 Jul, 2025 Reviewers invited by journal 23 Jul, 2025 Editor assigned by journal 18 Jul, 2025 Submission checks completed at journal 18 Jul, 2025 First submitted to journal 16 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-7136122","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":490568971,"identity":"be577f76-7b64-4161-9072-9b2dd59ec010","order_by":0,"name":"Yangyang Liu","email":"","orcid":"","institution":"Northwest University","correspondingAuthor":false,"prefix":"","firstName":"Yangyang","middleName":"","lastName":"Liu","suffix":""},{"id":490568975,"identity":"9e94de16-cab8-4b9b-b4cd-40d6732e00ca","order_by":1,"name":"Jiatong Liu","email":"","orcid":"","institution":"Northwest University","correspondingAuthor":false,"prefix":"","firstName":"Jiatong","middleName":"","lastName":"Liu","suffix":""},{"id":490568977,"identity":"ca3891fb-9b29-44da-9421-849449febc88","order_by":2,"name":"Xinda Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxElEQVRIiWNgGAWjYBACPmYgkcDAzMAP4TMT1sIG0yLZQLQWmOEGB4jWws5jJvFwh7W88e0eMwmGCuvEBvazBwg4jMfYIPFMuuG2O2eAWs6kJzbw5CUQ0mL4ILHtMOO2G7nbJBjbDic2SPAYENJicACoxX7zDJCWf8RpAduSuEECpKWBKC1sxQaJbenJM27kf7ZIOJZu3MaTg18LP//hbZI/26xt+2ekJd74UGMt289+Br8WVJDAAI+pUTAKRsEoGAWUAABIgDyBnqS97QAAAABJRU5ErkJggg==","orcid":"","institution":"Northwest University","correspondingAuthor":true,"prefix":"","firstName":"Xinda","middleName":"","lastName":"Liu","suffix":""},{"id":490568979,"identity":"66428dd5-b792-4198-bbf2-83695ca9f031","order_by":3,"name":"Guohua Geng","email":"","orcid":"","institution":"Northwest University","correspondingAuthor":false,"prefix":"","firstName":"Guohua","middleName":"","lastName":"Geng","suffix":""},{"id":490568981,"identity":"af020f65-4fda-4664-bff1-8f15c55273c7","order_by":4,"name":"Zhan Li","email":"","orcid":"","institution":"Northwest University","correspondingAuthor":false,"prefix":"","firstName":"Zhan","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2025-07-16 05:53:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7136122/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7136122/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s40494-026-02391-0","type":"published","date":"2026-02-24T15:58:10+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":103765546,"identity":"926ba6cb-3b01-43ac-bc22-af0ff74a912c","added_by":"auto","created_at":"2026-03-02 16:03:53","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3624744,"visible":true,"origin":"","legend":"","description":"","filename":"InSwAVInvolutionenhancedfeatureclusteringandswappedassignmentsforporcelainrelicmicroscopicimageclassification.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7136122/v1_covered_0dbc5444-c56f-4575-b806-88518c404b63.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"InSwAV: Involution enhanced feature clustering and swapped assignments for porcelain relic microscopic image classification","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":"npj-heritage-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"hsci","sideBox":"Learn more about [Heritage Science](http://heritagesciencejournal.springeropen.com)","snPcode":"40494","submissionUrl":"https://submission.nature.com/new-submission/40494/3","title":"npj Heritage Science","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7136122/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7136122/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe precise classification of porcelain relic fragments is a crucial step in the process of artifact restoration and reconstruction. The current classification methods based on apparent features such as glaze color and patterns can achieve preliminary classification, but they have obvious limitations when dealing with porcelain fragments with highly similar visual features. Especially when the sample size of microscopic images is limited, the classification performance of traditional methods often drops significantly. To address the challenges of high computational complexity and low efficiency in traditional contrastive learning for image classification, we propose InSwAV, which aligns deep features with cluster prototypes by swapped assignment through involution-enhanced feature extraction and cross-view consistency enforcement. First, we propose ResInv combining involution blocks with residual blocks of the feature extraction network. Next, features extracted by ResInv are clustered, and consistency in cluster assignments is enforced across multi-view augmented samples via cross-prediction mechanisms. This approach iteratively optimizes feature representation by minimizing cross-entropy loss, significantly reducing training time while maintaining accuracy. Furthermore, we constructed the Porcelain Relic Microscopic Images(PRMI) dataset of five classification, with data augmentation applied to enhance model robustness. Experimental results show that InSwAV achieves a classification accuracy of 96.2% on the porcelain relic microscopic image dataset, outperforming existing methods.\u003c/p\u003e","manuscriptTitle":"InSwAV: Involution enhanced feature clustering and swapped assignments for porcelain relic microscopic image classification","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-25 08:01:24","doi":"10.21203/rs.3.rs-7136122/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-16T01:33:04+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-13T14:30:31+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-13T07:56:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-25T14:21:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"140342781066393432470402790679677229883","date":"2025-07-25T14:14:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"61283350362027877288793811688497511467","date":"2025-07-25T09:25:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"161169385905573387492404103095053940388","date":"2025-07-25T01:10:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"83106641486217331566959546056890472155","date":"2025-07-24T01:55:29+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-23T09:12:36+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-18T10:01:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-18T09:57:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj heritage science","date":"2025-07-16T05:37:38+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-heritage-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"hsci","sideBox":"Learn more about [Heritage Science](http://heritagesciencejournal.springeropen.com)","snPcode":"40494","submissionUrl":"https://submission.nature.com/new-submission/40494/3","title":"npj Heritage Science","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7682c067-163e-432e-8aa2-901c8932f103","owner":[],"postedDate":"July 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-02T16:01:17+00:00","versionOfRecord":{"articleIdentity":"rs-7136122","link":"https://doi.org/10.1038/s40494-026-02391-0","journal":{"identity":"npj-heritage-science","isVorOnly":false,"title":"npj Heritage Science"},"publishedOn":"2026-02-24 15:58:10","publishedOnDateReadable":"February 24th, 2026"},"versionCreatedAt":"2025-07-25 08:01:24","video":"","vorDoi":"10.1038/s40494-026-02391-0","vorDoiUrl":"https://doi.org/10.1038/s40494-026-02391-0","workflowStages":[]},"version":"v1","identity":"rs-7136122","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7136122","identity":"rs-7136122","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}