Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{206692,
        author = {Ajay U Bangera and Sahana Bandekar and C Jeevesh and Sharanya and Spoorthi},
        title = {Real-Time Abnormal Activity Detection and Alert System Using Computer Vision},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {no},
        pages = {230-233},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=206692},
        abstract = {This paper presents a novel, multi-tenant Software-as-a-Service (SaaS) computer vision framework for the automated detection of abnormal behaviors. Historically, video surveillance has relied on manual observation or basic motion detection, which is labor-intensive and prone to severe false-positive alarm rates. Our proposed system addresses these inefficiencies by employing a dual-tier AI architecture. The system integrates YOLOv8 for high-speed, real-time anomaly detection at the edge, coupled with a secondary verification layer utilizing a Vision-Language Model (Gemini Vision API) to analyze snapshots and eliminate false alarms. Hosted on a multi-tenant architecture, the platform allows users to create isolated projects (e.g., for Educational Institutions or Government Systems), manage varied CCTV inputs, and enforce Role-Based Access Control (RBAC). The system logs verified alerts to a relational database and autonomously dispatches email notifications to authorized members. This demonstrates a scalable, highly accurate solution for proactive security deployments.},
        keywords = {Abnormal Activity Detection, Multi-Tenant Architecture, YOLOv8, Vision-Language Models, False Alarm Mitigation, Video Surveillance.},
        month = {July},
        }