AI-Driven Visualization of CCTV Data for Enhanced School Violence Monitoring and Response
Phuong Anh Nguyen, Le Anh Ngoc · 2025
Effectively addressing school violence is essential for maintaining a safe learning environment, yet existing systems for visualizing detected anomalies, such as physical altercations, remain limited. A study indicates that over 20% of students in the U.S. have experienced violence or bullying on school grounds, highlighting the urgent need for more advanced monitoring and response systems. While current technologies can detect violent behavior using CCTV footage, schools often face challenges in making this data actionable due to a lack of clear visualizations. This paper presents an AI-driven solution that transforms raw detection data into intuitive visual formats—such as time-series charts, bar charts, and detailed tables—that include key information like timestamps, anomaly types, and direct links to CCTV footage. These visualizations help administrators recognize patterns, such as recurring incidents at specific times or locations, enabling more effective real-time interventions and improving overall school safety.