• Sanjana .V. Shanbhag
• Srushti D. Morade
• Deepak Ranoji Naik

In modern environments, ensuring safety through ef- fective surveillance has become very essential. Old, traditional monitoring systems depend a lot on human observation, which may result in delayed responses and overlooked incidents. To overcome these limitations, SafeEYE AI (Artificial Intelligence) introduces an automated surveillance solution which is designed to detect high-risk anomalies such as fire, smoke, and fights in real time using CCTV (Closed-Circuit Television) footage. The SafeEYE AI system integrates advanced computer vision and deep learning methods to detect abnormal patterns from video streams. A backend server processes CCTV inputs and sends frames to the trained AI model, which classifies activities and detects anomalies using thresholding. Events which exceed a certain threshold, trigger instant alerts via Firebase Cloud Messaging (FCM), which delivers both visual and textual evi- dence to the dashboard. The scalable design of SafeEYE AI helps with easy integration, accurate detection, and quick notification delivery. This system, SafeEYE AI, improves awareness, lessens dependency on people, and fortifies public safety by combining automation, intelligence, and real-time decision support.

Surveillance, Anomaly Detection, Computer Vision, Real-Time Monitoring, Public Safety

"SafeEYE AI: AI-Powered CCTV Surveillance for Real-Time Anomaly Detection with Instant Notifications", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 11, page no.b657-b659, November-2025, Available :http://www.jetir.org/papers/JETIR2511181.pdf

"SafeEYE AI: AI-Powered CCTV Surveillance for Real-Time Anomaly Detection with Instant Notifications", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 11, page no. ppb657-b659, November-2025, Available at : http://www.jetir.org/papers/JETIR2511181.pdf