Yolo-Based Anomaly Detection & Alert System
Yolo-Based Anomaly Detection & Alert System
Ms. R D Vidyarani1, Chinmay B H2 Dheeraj A3, Dheeraj M S4, Harish Deekshit5
Assistant Professor, Dept of CSE, KSIT, Karnataka, India1
Student, Dept of CSE, KSIT, Karnataka, India2-5
ABSTRACT – ATM security has become an important concern because these locations are often exposed to theft, vandalism, unauthorized access, and other suspicious activities that may threaten public safety and financial security. Continuous manual monitoring of CCTV cameras is difficult and may not always ensure a quick response during an incident. To address this issue, this project proposes a real-time YOLO-based suspicious activity detection and alert system for ATM environments. The system is designed to analyze video frames from CCTV footage, detect human presence and unusual behavior, and generate alerts when suspicious activity is identified. By using an object detection approach, the proposed system aims to improve the speed and accuracy of threat recognition while reducing the need for constant human supervision.The model can support better surveillance, quicker intervention, and stronger protection of ATM premises. Overall, this project presents a practical and efficient method for enhancing ATM safety through automated real-time monitoring and alert generation.KEYWORDS: YOLO, ATM security, suspicious activity detection, real-time surveillance, CCTV monitoring, alert generation, object detection, theft prevention