Paper Title
Real-Time Violence Detection and Alert System Using MobilenetV2 and Cloud Fire Store
Abstract
The aggressive behaviors pose a major threat to personal safety and quality of life. Many other approaches have
been taken to stop the aggressive behavior, including setting up and maintaining surveillance systems. If the surveillance
systems can automatically identify violent behavior and provide warning or alarm signals to the police right away, it will be
of tremendous relevance. Computer vision and image processing have become active study areas for abnormal and violent
activity detection, attracting new researchers. This paper proposes a real-time violence detection and alert system that
employs MobileNetV2 for violence detection in video streams. Leveraging Cloud Firestore for data storage and management
and a Telegram Bot for instant alerting, the system provides a scalable, accurate, and adaptable solution for public safety.
Keywords - Violence, Violent Activity, Real time, Violent Behavior, Surveillance, Violence Detection, Alert, Storage