Performance Analysis of the YOLO object detection algorithm in embedded systems: generated code vs. native implementation

Performance Analysis of the YOLO object detection algorithm in embedded systems: generated code vs. native implementation | 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 Performance Analysis of the YOLO object detection algorithm in embedded systems: generated code vs. native implementation Pablo Martínez Otero, Mar Hernández Melero, Alberto Tellaeche Iglesias, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7678531/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 This work presents a comparative evaluation of advanced YOLO architectures for object detection, with a specific focus on their performance in traffic light detection for autonomous driving applications. Two deployment strategies were analyzed: a native implementation using PyTorch and a Model Based Engineering (MBE) implementation through automatic code generation. Evaluation metrics included precision-recall curves and confusion matrices across varying Intersection over Union (IoU) thresholds, as well as mean Average Precision (mAP) to assess detection quality and inference time measurements to evaluate computational efficiency on embedded platforms. The evaluation was based on a custom video extracted from the CARLA simulator, which was meticulously annotated by reviewing each frame to ensure the accuracy of the labeling. The study highlights the compromises between model accuracy and computational cost, providing a reproducible framework for performance benchmarking of object detection algorithms in safety-critical environments. YOLO deep learning embedded systems autonomous driving traffic light detection code generation Full Text Additional Declarations No competing interests reported. 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-7678531","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":533961025,"identity":"d2438f4d-cedc-40e9-ab5d-6d4eaed70299","order_by":0,"name":"Pablo Martínez Otero","email":"data:image/png;base64,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","orcid":"","institution":"Embedded Systems, IKERLAN","correspondingAuthor":true,"prefix":"","firstName":"Pablo","middleName":"Martínez","lastName":"Otero","suffix":""},{"id":533961026,"identity":"1e44fc39-f46c-4ca0-affe-fc60cf8e494b","order_by":1,"name":"Mar Hernández Melero","email":"","orcid":"","institution":"Embedded Systems, IKERLAN","correspondingAuthor":false,"prefix":"","firstName":"Mar","middleName":"Hernández","lastName":"Melero","suffix":""},{"id":533961027,"identity":"24ff7a9c-e20b-450f-be5a-1e75b524c202","order_by":2,"name":"Alberto Tellaeche Iglesias","email":"","orcid":"","institution":"University of Deusto","correspondingAuthor":false,"prefix":"","firstName":"Alberto","middleName":"Tellaeche","lastName":"Iglesias","suffix":""},{"id":533961028,"identity":"726d8d49-53e8-4acc-842e-481db9ac6187","order_by":3,"name":"Jon Pedernales Pereira","email":"","orcid":"","institution":"Embedded Systems, IKERLAN","correspondingAuthor":false,"prefix":"","firstName":"Jon","middleName":"Pedernales","lastName":"Pereira","suffix":""}],"badges":[],"createdAt":"2025-09-22 14:24:23","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7678531/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7678531/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94549505,"identity":"cbd24195-1d9c-4baa-afe5-3f174bf55016","added_by":"auto","created_at":"2025-10-28 17:46:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1776782,"visible":true,"origin":"","legend":"","description":"","filename":"PabloMartinezOteroSpringerNaturev2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7678531/v1/4074eba79699e5111d14cfec.pdf"},{"id":94549500,"identity":"4bfe43de-7877-4e19-9ca6-810022208d90","added_by":"auto","created_at":"2025-10-28 17:46:01","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5542,"visible":true,"origin":"","legend":"","description":"","filename":"5c3bb10bde19481395e24bfb14535d4f.json","url":"https://assets-eu.researchsquare.com/files/rs-7678531/v1/aaba7ab727eba907ebbe684b.json"},{"id":95224769,"identity":"3e284e51-587e-41df-8dda-32e7a23e9868","added_by":"auto","created_at":"2025-11-05 16:24:16","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1722035,"visible":true,"origin":"","legend":"","description":"","filename":"PabloMartinezOteroSpringerNaturev2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7678531/v1_covered_da736233-e597-4fd7-969d-92f274ac0f2d.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Performance Analysis of the YOLO object detection algorithm in embedded systems: generated code vs. native implementation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"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":"YOLO, deep learning, embedded systems, autonomous driving, traffic light detection, code generation","lastPublishedDoi":"10.21203/rs.3.rs-7678531/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7678531/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"This work presents a comparative evaluation of advanced YOLO architectures for object detection, with a specific focus on their performance in traffic light detection for autonomous driving applications. Two deployment strategies were analyzed: a native implementation using PyTorch and a Model Based Engineering (MBE) implementation through automatic code generation. Evaluation metrics included precision-recall curves and confusion matrices across varying Intersection over Union (IoU) thresholds, as well as mean Average Precision (mAP) to assess detection quality and inference time measurements to evaluate computational efficiency on embedded platforms. The evaluation was based on a custom video extracted from the CARLA simulator, which was meticulously annotated by reviewing each frame to ensure the accuracy of the labeling. The study highlights the compromises between model accuracy and computational cost, providing a reproducible framework for performance benchmarking of object detection algorithms in safety-critical environments.","manuscriptTitle":"Performance Analysis of the YOLO object detection algorithm in embedded systems: generated code vs. native implementation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-28 03:17:45","doi":"10.21203/rs.3.rs-7678531/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":"7be75c44-4349-4f1b-8b80-9cc8fd6e80f5","owner":[],"postedDate":"October 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-04T08:39:03+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-28 03:17:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7678531","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7678531","identity":"rs-7678531","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}