TaxonomyNet: A Consistent and Efficient Model for Taxonomic Rank Identification in Wildlife Images

TaxonomyNet: A Consistent and Efficient Model for Taxonomic Rank Identification in Wildlife Images | 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 TaxonomyNet: A Consistent and Efficient Model for Taxonomic Rank Identification in Wildlife Images Qianqian Zhang, Khandakar Ahmed, Chenhao Xu, Muhammad Khan, Hua Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7173596/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 21 Jan, 2026 Read the published version in Scientific Reports → Version 1 posted 15 You are reading this latest preprint version Abstract Accurate and scalable taxonomic classification is essential for biodiversity research, supporting systematic species identification across multiple hierarchical ranks. However, current image-based classification methods often fail to enforce taxonomic consistency, limiting their utility in ecological monitoring and species discovery. Additionally, field-based biodiversity studies are constrained by limited computational resources and network availability on edge devices. To address the above challenges, this paper proposes an ensemble detection model, TaxonomyNet, that integrates six independent heads corresponding to different taxonomic classifications. To improve prediction consistency across taxonomic ranks, we introduce the Weighted Agreement Loss (WAL) metric—a confidence-weighted disagreement measure designed to quantify and minimise inconsistencies between predicted outputs and a reference taxonomy. TaxonomyNet achieves high detection performance across all ranks (mAP: 90.7%–99.75%) after training on a dataset of 50 Australian animal species with taxonomic classification labels. Compared with both lightweight and large-scale foundation models, the proposed method improves hierarchical classification reliability (by up to 3.87%) and computational efficiency (reducing delay by 22 minutes across 1,500 samples) on edge devices. This work provides a practical and extensible solution for hierarchical classification in real-world biodiversity monitoring scenarios. Biological sciences/Computational biology and bioinformatics Biological sciences/Ecology Earth and environmental sciences/Ecology Full Text Additional Declarations Competing interest reported. We would like to disclose that one of the co-authors of this manuscript, Professor Hua Wang, is a member of the Scientific Reports Editorial Board. However, we confirm that there have been no prior discussions with Professor Wang or any other Editorial Board Member regarding the content or submission of this manuscript. All co-authors, including Professor Wang, have adhered to the journal’s policies to ensure a transparent and unbiased submission process. Supplementary Files SRsupplementalfileApp1.pdf SRsupplementalfileApp2.pdf SRsupplementalfileApp3.pdf Cite Share Download PDF Status: Published Journal Publication published 21 Jan, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 29 Aug, 2025 Reviewers agreed at journal 26 Aug, 2025 Reviewers agreed at journal 24 Aug, 2025 Reviewers agreed at journal 23 Aug, 2025 Reviews received at journal 22 Aug, 2025 Reviews received at journal 19 Aug, 2025 Reviewers agreed at journal 07 Aug, 2025 Reviewers agreed at journal 07 Aug, 2025 Reviewers agreed at journal 04 Aug, 2025 Reviewers agreed at journal 01 Aug, 2025 Reviewers invited by journal 01 Aug, 2025 Editor assigned by journal 01 Aug, 2025 Editor invited by journal 30 Jul, 2025 Submission checks completed at journal 28 Jul, 2025 First submitted to journal 28 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-7173596","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":498789020,"identity":"4acbdb3d-5aed-4b5e-b52b-b22b0686b34e","order_by":0,"name":"Qianqian Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAUlEQVRIiWNgGAWjYDACZhBhACJ4QIQNWFCCFC1pRGhBALCWw4S18B1nfvbgTYFdHgP/2YOfC36dT9xwgPngbR4Gu8QGHFokD7OZG84xSC5mYDiXLD2z7zZQC1uyNQ9DMk4tBocZzKR5DJgTGxh7DKR5e24bGxzgAYowMOPRwv4NqKU+sYGZx/g3b885oBZ+oAhDPR4tIDMNDic2sIEYPw7IAW1hA2o5jMcvPGWScwyOJ7bx8JhZ8zYkywF9Z2wJFDHGpYXv/PFtEm/+VCf2858xvs3zx46H73jzwxtvKqplcWlhOMAAiRE2EIexjQE5cglogYA/eFSOglEwCkbBiAUAy5hOmnVyrm8AAAAASUVORK5CYII=","orcid":"","institution":"Victoria University","correspondingAuthor":true,"prefix":"","firstName":"Qianqian","middleName":"","lastName":"Zhang","suffix":""},{"id":498789021,"identity":"fb73575c-6c1e-431a-af02-9d68fab300ff","order_by":1,"name":"Khandakar Ahmed","email":"","orcid":"","institution":"Victoria University","correspondingAuthor":false,"prefix":"","firstName":"Khandakar","middleName":"","lastName":"Ahmed","suffix":""},{"id":498789028,"identity":"b164fb23-08e3-488d-9bbd-0fa20f7ff7df","order_by":2,"name":"Chenhao Xu","email":"","orcid":"","institution":"Victoria University","correspondingAuthor":false,"prefix":"","firstName":"Chenhao","middleName":"","lastName":"Xu","suffix":""},{"id":498789029,"identity":"d082ef4a-b295-490f-8021-05bc4d198593","order_by":3,"name":"Muhammad Khan","email":"","orcid":"","institution":"Victoria University","correspondingAuthor":false,"prefix":"","firstName":"Muhammad","middleName":"","lastName":"Khan","suffix":""},{"id":498789030,"identity":"c193c3bf-dfdb-4a53-98ff-29d6b3c38e0a","order_by":4,"name":"Hua Wang","email":"","orcid":"","institution":"Victoria University","correspondingAuthor":false,"prefix":"","firstName":"Hua","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-07-21 05:53:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7173596/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7173596/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-34944-x","type":"published","date":"2026-01-21T15:56:54+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":101151719,"identity":"40b1dac0-41d9-4c4b-9aac-bac0db15887c","added_by":"auto","created_at":"2026-01-26 16:02:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1184674,"visible":true,"origin":"","legend":"","description":"","filename":"mainv2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7173596/v1_covered_5e2c3f74-04df-47d7-95ae-fbbaf87da67c.pdf"},{"id":88847733,"identity":"2761a819-56d7-4ac6-89da-eb58f4ba541c","added_by":"auto","created_at":"2025-08-12 04:29:42","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":63353,"visible":true,"origin":"","legend":"","description":"","filename":"SRsupplementalfileApp1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7173596/v1/75512231f3e1add35bd10db9.pdf"},{"id":88847730,"identity":"47549aaf-8c86-4a03-a230-2f1637867ce3","added_by":"auto","created_at":"2025-08-12 04:29:42","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":64069,"visible":true,"origin":"","legend":"","description":"","filename":"SRsupplementalfileApp2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7173596/v1/3b4e05ee6c771532b5a4a39a.pdf"},{"id":88848062,"identity":"373d4787-4722-4960-8799-84074542f777","added_by":"auto","created_at":"2025-08-12 04:37:42","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":63159,"visible":true,"origin":"","legend":"","description":"","filename":"SRsupplementalfileApp3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7173596/v1/995349a6191d9eb264ac664f.pdf"}],"financialInterests":"Competing interest reported. We would like to disclose that one of the co-authors of this manuscript, Professor Hua Wang, is a member of the Scientific Reports Editorial Board. However, we confirm that there have been no prior discussions with Professor Wang or any other Editorial Board Member regarding the content or submission of this manuscript. All co-authors, including Professor Wang, have adhered to the journal’s policies to ensure a transparent and unbiased submission process.","formattedTitle":"TaxonomyNet: A Consistent and Efficient Model for Taxonomic Rank Identification in Wildlife Images","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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7173596/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7173596/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Accurate and scalable taxonomic classification is essential for biodiversity research, supporting systematic species identification across multiple hierarchical ranks. However, current image-based classification methods often fail to enforce taxonomic consistency, limiting their utility in ecological monitoring and species discovery. Additionally, field-based biodiversity studies are constrained by limited computational resources and network availability on edge devices. To address the above challenges, this paper proposes an ensemble detection model, TaxonomyNet, that integrates six independent heads corresponding to different taxonomic classifications. To improve prediction consistency across taxonomic ranks, we introduce the Weighted Agreement Loss (WAL) metric—a confidence-weighted disagreement measure designed to quantify and minimise inconsistencies between predicted outputs and a reference taxonomy. TaxonomyNet achieves high detection performance across all ranks (mAP: 90.7%–99.75%) after training on a dataset of 50 Australian animal species with taxonomic classification labels. Compared with both lightweight and large-scale foundation models, the proposed method improves hierarchical classification reliability (by up to 3.87%) and computational efficiency (reducing delay by 22 minutes across 1,500 samples) on edge devices. This work provides a practical and extensible solution for hierarchical classification in real-world biodiversity monitoring scenarios.","manuscriptTitle":"TaxonomyNet: A Consistent and Efficient Model for Taxonomic Rank Identification in Wildlife Images","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-12 04:29:37","doi":"10.21203/rs.3.rs-7173596/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-29T17:40:48+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"215008187366792497493274557268885960708","date":"2025-08-26T11:23:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"158972725027465719625315962367326182182","date":"2025-08-25T02:38:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"144467139449795508287509914564019891954","date":"2025-08-23T07:51:53+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-22T21:19:39+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-19T06:01:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"201220092652592073995552331882650774898","date":"2025-08-07T09:10:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"3857805971420274832497899059916862144","date":"2025-08-07T08:29:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"231147856442095982003404364977020347993","date":"2025-08-05T00:30:18+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"249161163476823771707491082327403514335","date":"2025-08-01T16:31:19+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-01T12:42:22+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-01T12:38:00+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-07-30T09:17:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-29T03:02:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-07-29T03:00:43+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1e446448-26b4-416b-8825-539bcfe0f947","owner":[],"postedDate":"August 12th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":52976617,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":52976618,"name":"Biological sciences/Ecology"},{"id":52976619,"name":"Earth and environmental sciences/Ecology"}],"tags":[],"updatedAt":"2026-01-26T16:00:20+00:00","versionOfRecord":{"articleIdentity":"rs-7173596","link":"https://doi.org/10.1038/s41598-025-34944-x","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-01-21 15:56:54","publishedOnDateReadable":"January 21st, 2026"},"versionCreatedAt":"2025-08-12 04:29:37","video":"","vorDoi":"10.1038/s41598-025-34944-x","vorDoiUrl":"https://doi.org/10.1038/s41598-025-34944-x","workflowStages":[]},"version":"v1","identity":"rs-7173596","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7173596","identity":"rs-7173596","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}