A Review of Smart Atm Security System Using ESP32

Automated Teller Machines (ATMs) are integral to modern banking but are also prime targets for theft, vandalism, and tampering. Conventional CCTV-based systems mainly provide post-incident evidence, failing to offer real-time preventive measures. This paper reviews the design and development of a Smart ATM Security System using ESP32-CAM, integrating IoT, sensor fusion, and machine learning (ML) for proactive, autonomous threat detection and response. The system utilizes sensors such as PIR motion, vibration, smoke, and gas detectors to monitor the ATM environment, transmitting data via Wi-Fi to a backend server for analysis using models like YOLO or MobileNet. Depending on the classified threat level, automated actions— such as buzzer, foam spray, or ink spray activation—are triggered. This review summarizes the architecture, working methodology, advantages, limitations, and potential future enhancements of such IoT-based ATM security systems.