Lightweight Visual Detection Framework for Endangered Species Monitoring in Complex Natural Environments

Lightweight Visual Detection Framework for Endangered Species Monitoring in Complex Natural Environments | 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 Lightweight Visual Detection Framework for Endangered Species Monitoring in Complex Natural Environments Xinyao Yang, Junyi Chen, Nanqing Sun This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9677189/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 Wildlife monitoring in complex natural environments demands robust and efficient visual detection systems, especially for endangered species protection. Existing detection models suffer from low accuracy, high computational cost, and poor adaptability under low-light and cluttered backgrounds, limiting field deployment. This work proposes a lightweight detection framework optimized for real-time identification of wild Chinese giant salamanders. The method integrates an efficient multi-scale attention module, adopts a Wise-IoU loss for stable bounding box regression, and employs lightweight adaptive convolution to reduce computation. Experiments show the model achieves 94.85% recall, 95.39% precision, 95.12% F1 score, and 77.20 fps with only 11.56 MB memory. It outperforms baseline YOLO v11n significantly and has been validated in a national nature reserve with over 20 successful detections. This work provides a deployable visual solution for endangered species conservation and offers a transferable pipeline for visual computing in ecological monitoring. The data code mentioned above can be obtained from the following URL: https://doi.org/10.5281/zenodo.20116997 Machine learning YOLO11n Siamese network Chinese giant salamander Full Text Additional Declarations The authors declare no competing interests. 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-9677189","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":638085665,"identity":"080f26cb-8589-4c6c-a851-5801d2c4921d","order_by":0,"name":"Xinyao Yang","email":"","orcid":"","institution":"Hubei Minzu University","correspondingAuthor":false,"prefix":"","firstName":"Xinyao","middleName":"","lastName":"Yang","suffix":""},{"id":638085666,"identity":"de56c8c8-c65a-4d39-b066-53c6c13d0786","order_by":1,"name":"Junyi Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIie3RIQsCMRTA8QeD0zC0PlHOr7BDOPw4GyeXFGwuGBRlC2LXb2E0apll9rOdxazNIipXlbuzGfbL+/P2NgDH+UNeZXa4InvSNiH7lMtxcVKjJoauJH6gVcRSa4oTH/shSEs6YG3YOM9JiYu9k/SkPDFdxbEUEw/qesHzE3oZBWtFhcbIJGLbArTHTX5SiTbNhsJsSiKsBwwHBQlw1nwoJiZJPxwKRUok1V6IaHm2PpRLqIkYyl32yMitoYW7tPVsnyLbZV95u8uxX9fL/OQD/e244ziO89ULwwFMRrEhK40AAAAASUVORK5CYII=","orcid":"","institution":"Hubei Minzu University","correspondingAuthor":true,"prefix":"","firstName":"Junyi","middleName":"","lastName":"Chen","suffix":""},{"id":638085667,"identity":"65cc7fbc-430e-4d61-8b54-3ef3910c4b6e","order_by":2,"name":"Nanqing Sun","email":"","orcid":"","institution":"Hubei Minzu University","correspondingAuthor":false,"prefix":"","firstName":"Nanqing","middleName":"","lastName":"Sun","suffix":""}],"badges":[],"createdAt":"2026-05-11 09:08:07","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-9677189/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9677189/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109068018,"identity":"e91eeb4f-6a4e-4397-a8c5-b8325095a417","added_by":"auto","created_at":"2026-05-12 10:02:57","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":943038,"visible":true,"origin":"","legend":"","description":"","filename":"LightweightVisualDetectionFrameworkforEndangeredSpeciesMonitoringinComplexNaturalEnvironments.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9677189/v1_covered_dfd4fa9c-3021-43ef-9fd3-344d67aeba3c.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eLightweight Visual Detection Framework for Endangered Species Monitoring in Complex Natural Environments\u003c/strong\u003e\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Hubei Minzu University","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":"Machine learning, YOLO11n, Siamese network, Chinese giant salamander","lastPublishedDoi":"10.21203/rs.3.rs-9677189/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9677189/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWildlife monitoring in complex natural environments demands robust and efficient visual detection systems, especially for endangered species protection. Existing detection models suffer from low accuracy, high computational cost, and poor adaptability under low-light and cluttered backgrounds, limiting field deployment. This work proposes a lightweight detection framework optimized for real-time identification of wild Chinese giant salamanders. The method integrates an efficient multi-scale attention module, adopts a Wise-IoU loss for stable bounding box regression, and employs lightweight adaptive convolution to reduce computation. Experiments show the model achieves 94.85% recall, 95.39% precision, 95.12% F1 score, and 77.20 fps with only 11.56 MB memory. It outperforms baseline YOLO v11n significantly and has been validated in a national nature reserve with over 20 successful detections. This work provides a deployable visual solution for endangered species conservation and offers a transferable pipeline for visual computing in ecological monitoring. The data code mentioned above can be obtained from the following URL: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.20116997\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.20116997\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e","manuscriptTitle":"Lightweight Visual Detection Framework for Endangered Species Monitoring in Complex Natural Environments","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-12 03:36:34","doi":"10.21203/rs.3.rs-9677189/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":"21986e9f-7b32-4e60-8b7e-142d7f79bc44","owner":[],"postedDate":"May 12th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-12T03:36:34+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-12 03:36:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9677189","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9677189","identity":"rs-9677189","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}