Real-Time Violence Detection and Alert System using MobileNetV2 and Cloud Firestore
Mannu Thomas, P. Balamurugan · 2024
The aggressive action greatly jeopardizes individual safety and general well-being. Various alternative tactics have been used to reduce violent behavior, including the installation and maintenance of surveillance systems. It is crucial for monitoring systems to automatically detect aggressive behavior and promptly provide warning signals to the authorities. Computer vision and image processing have emerged as crucial study areas for identifying deviant and violent behavior, drawing in new researchers. The paper introduces a real- time violence detection and alert system that identifies violence in video streams by utilizing MobileNetV2. The system provides a scalable, precise, and adaptable public safety solution by using Cloud Firestore for data storage and administration, and a Telegram Bot for instant notifications.