Next-Generation Public Safety System for Real-Time Crowd and Weapon Detection using YOLOv11 with Automated Alerts

Safety security in the crowded public environment is both important and difficult to achieve, as the typical surveillance systems provide almost no real-time identification of violent actions so the reaction to possible dangers is delayed. The proposed system introduces top-notch surveillance method that uses YOLOv11 model for object detection, Optical Flow for detecting sudden motion, and LSTM networks for analyzing motions sequences. This combination allows for capturing the density of the crowds and any violence they intend to cause; the systems also alert the security personnel when it has detected any violators. Proving effectiveness in tests up to 98.7% and permitting a proper detection and prevention of threats in public safety, this integrated system allows for risk elimination in large-population places and serves as a proactive system rather than the reactive system that current monitoring method.