{"paper_id":"02b44fcf-42e3-4a75-89df-efe68235fd22","body_text":"Violence Detection from Industrial Surveillance Videos Using Ensemble Learning | 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 Violence Detection from Industrial Surveillance Videos Using Ensemble Learning Hamza Khan, Kenneth Eze, Xiaohong Yuan, Letu Qingge, Kaushik Roy This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9237080/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 6 You are reading this latest preprint version Abstract The detection of violence in videos has emerged as a critical application for enhancing public safety and security in real-world scenarios. Despite significant advancements in the field, challenges persist due to the inherent complexity of video data and the sudden, unpredictable nature of violent actions. Achieving consistent performance across diverse and advanced real-life datasets remains a formidable task. To address these challenges, this paper proposes a novel stacking ensemble-based approach that integrates three pretrained models with a stacked 3D Convolutional Neural Network (CNN) architecture. This design aims to enhance the performance of individual models while leveraging the strengths of ensemble learning. The proposed method is rigorously evaluated on three publicly available datasets—RLVS, RWF-2000, and Hockey Fight—achieving remarkable results. Specifically, the model attains accuracy rates of 96.5% on RLVS, 97.0% on RWF-2000, and 97.5% on Hockey Fight, alongside F1-scores of 96.5% on RLVS, 97.0% on RWF-2000, and 97.5% on Hockey Fight. Furthermore, cross-dataset analysis demonstrates the model’s robust generalization capabilities. For instance, when trained on RLVS and tested on RWF-2000, the model achieves an accuracy of 78.5%. Similarly, training on RWF-2000 and testing on Hockey Fight yields an accuracy of 81.2%, while training on Hockey Fight and testing on RLVS results in an accuracy of 76.8%. These findings underscore the model’s potential for real-world industrial surveillance applications, where adaptability to diverse environments is crucial. Violence Detection Activity detection Industrial surveillance Ensemble Learning Computer Vision Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 04 May, 2026 Reviewers invited by journal 04 May, 2026 Editor assigned by journal 30 Apr, 2026 Editor invited by journal 30 Apr, 2026 Submission checks completed at journal 28 Apr, 2026 First submitted to journal 28 Apr, 2026 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-9237080\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":634328608,\"identity\":\"c48bdf57-35a3-42fe-b101-7c34ca4da566\",\"order_by\":0,\"name\":\"Hamza Khan\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/UlEQVRIiWNgGAWjYBACAx4eIFlhAyQYGxDCPAS1nElD0sJGjBbGtsNIQoS0mPOcPfi5gO28vLz/4saHPxjuyPPPb2B88LYNtxbL3r5k6Rk8tw033njYbMzD8MxwxjEGZsO5eLQYnOcxkOaRuM24ccbBNmkGhsMJDMcY2KR58Wsx/s1jcM4eqKX95w+gFvljDOy/8Wo522MmzZNwIHE+f2Mb0NeHEwyAtjDj1XLmjJk1z4Hk5A0SjM3SPAaHDTceS2yWnHMOn5Yc49u8/+xs5/cff/jxR8VhebnDhw9+eFOGWwtC740EEAliIicDfEC+/wBxCkfBKBgFo2DkAQCF3VN0H0LdBQAAAABJRU5ErkJggg==\",\"orcid\":\"\",\"institution\":\"North Carolina Agricultural and Technical State University\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Hamza\",\"middleName\":\"\",\"lastName\":\"Khan\",\"suffix\":\"\"},{\"id\":634328610,\"identity\":\"502c20ff-87bd-47a3-8b85-ca8dfcc301a2\",\"order_by\":1,\"name\":\"Kenneth Eze\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Kenneth\",\"middleName\":\"\",\"lastName\":\"Eze\",\"suffix\":\"\"},{\"id\":634328612,\"identity\":\"c3427bef-ce7b-4533-85db-573bf1a1fc37\",\"order_by\":2,\"name\":\"Xiaohong Yuan\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"North Carolina Agricultural and Technical State University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Xiaohong\",\"middleName\":\"\",\"lastName\":\"Yuan\",\"suffix\":\"\"},{\"id\":634328615,\"identity\":\"a386cb6b-edea-47a5-9daa-cb63c471638a\",\"order_by\":3,\"name\":\"Letu Qingge\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"North Carolina Agricultural and Technical State University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Letu\",\"middleName\":\"\",\"lastName\":\"Qingge\",\"suffix\":\"\"},{\"id\":634328618,\"identity\":\"3f9dfa51-fcaa-4a49-acef-6b1948a6ecbf\",\"order_by\":4,\"name\":\"Kaushik Roy\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"North Carolina Agricultural and Technical State University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Kaushik\",\"middleName\":\"\",\"lastName\":\"Roy\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2026-03-26 17:54:01\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-9237080/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-9237080/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":109061850,\"identity\":\"63177998-c95d-4a32-bb4a-ccb2f0207734\",\"added_by\":\"auto\",\"created_at\":\"2026-05-12 08:34:49\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1465993,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"ViolenceDetectionfromIndustrialSurveillanceVideosUsingEnsembleLearning.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9237080/v1_covered_5f87bded-35cc-4db1-ae0e-f013757cb62e.pdf\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Violence Detection from Industrial Surveillance Videos Using Ensemble Learning\",\"fulltext\":[],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":false,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":true,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":true,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"discover-artificial-intelligence\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"diai\",\"sideBox\":\"Learn more about [Discover Artificial Intelligence](https://www.springer.com/44163)\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Discover Artificial Intelligence\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"stoa\",\"reportingPortfolio\":\"Discover Series\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Violence Detection, Activity detection, Industrial surveillance, Ensemble Learning, Computer Vision\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-9237080/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-9237080/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003eThe detection of violence in videos has emerged as a critical application for enhancing public safety and security in real-world scenarios. Despite significant advancements in the field, challenges persist due to the inherent complexity of video data and the sudden, unpredictable nature of violent actions. Achieving consistent performance across diverse and advanced real-life datasets remains a formidable task. To address these challenges, this paper proposes a novel stacking ensemble-based approach that integrates three pretrained models with a stacked 3D Convolutional Neural Network (CNN) architecture. This design aims to enhance the performance of individual models while leveraging the strengths of ensemble learning. The proposed method is rigorously evaluated on three publicly available datasets\\u0026mdash;RLVS, RWF-2000, and Hockey Fight\\u0026mdash;achieving remarkable results. Specifically, the model attains accuracy rates of 96.5% on RLVS, 97.0% on RWF-2000, and 97.5% on Hockey Fight, alongside F1-scores of 96.5% on RLVS, 97.0% on RWF-2000, and 97.5% on Hockey Fight. Furthermore, cross-dataset analysis demonstrates the model\\u0026rsquo;s robust generalization capabilities. For instance, when trained on RLVS and tested on RWF-2000, the model achieves an accuracy of 78.5%. Similarly, training on RWF-2000 and testing on Hockey Fight yields an accuracy of 81.2%, while training on Hockey Fight and testing on RLVS results in an accuracy of 76.8%. These findings underscore the model\\u0026rsquo;s potential for real-world industrial surveillance applications, where adaptability to diverse environments is crucial.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Violence Detection from Industrial Surveillance Videos Using Ensemble Learning\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2026-05-12 08:34:32\",\"doi\":\"10.21203/rs.3.rs-9237080/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"reviewerAgreed\",\"content\":\"81960070076447170740217239691070240238\",\"date\":\"2026-05-04T15:37:40+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewersInvited\",\"content\":\"\",\"date\":\"2026-05-04T10:12:33+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2026-04-30T09:00:03+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorInvited\",\"content\":\"\",\"date\":\"2026-04-30T08:19:02+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2026-04-28T21:34:30+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"Discover Artificial Intelligence\",\"date\":\"2026-04-28T21:31:48+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"discover-artificial-intelligence\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"diai\",\"sideBox\":\"Learn more about [Discover Artificial Intelligence](https://www.springer.com/44163)\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Discover Artificial Intelligence\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"stoa\",\"reportingPortfolio\":\"Discover Series\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"211efc8f-1c49-46b2-954c-1db30affb853\",\"owner\":[],\"postedDate\":\"May 12th, 2026\",\"published\":true,\"recentEditorialEvents\":[{\"type\":\"reviewerAgreed\",\"content\":\"81960070076447170740217239691070240238\",\"date\":\"2026-05-04T15:37:40+00:00\",\"index\":32,\"fulltext\":\"\"},{\"type\":\"reviewersInvited\",\"content\":\"9\",\"date\":\"2026-05-04T10:12:33+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2026-04-30T09:00:03+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorInvited\",\"content\":\"\",\"date\":\"2026-04-30T08:19:02+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"under-review\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2026-05-12T08:34:32+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2026-05-12 08:34:32\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-9237080\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-9237080\",\"identity\":\"rs-9237080\",\"version\":[\"v1\"]},\"buildId\":\"XKTyCvWXoU3ODBz1xrDgd\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}