Enhancing Public Safety: Real-Time Violence Detection and Notification System

The increasing incidence of violent events in public and private spaces necessitates the development of advanced security measures. This work introduces a novel Real-Time Violence Detection and Notification System, leveraging Faster R-CNN and MobileNetV2 architectures to analyze surveillance footage effectively. It enhances public safety by providing real-time alerts to security personnel and authorities via a Telegram bot, ensuring swift and efficient intervention. The system utilizes a combination of machine learning algorithms and computer vision techniques to accurately identify aggressive behavior while distinguishing it from non-violent activities. Comprehensive evaluations highlight the system’s high accuracy, and a comparative analysis demonstrates superior performance over existing models in detecting violent behavior.