Intelligent ATM Ingress Protection: “Integrating Facial Recognition and Liveness Dection for Enhanced the Security”

Quick cash withdrawal from ATMs is convenient, but it also increases the risks of theft and robbery in low-security areas. Fraud, impersonation, and unauthorized access to ATMs require a secure system. In this paper, we propose an IoT-based intelligent ATM access system with facial recognition and eye-blink detection to achieve secure transactions and high security. It checks users' identities before allowing them access to an ATM, so that multiple users or facial coverings cannot enter without permission. A liveness detection module confirms whether users blink, and distinguishes them from face masks or static images. Experimental results show that this approach outperforms RFID access and PIN-based authentication. By adding liveness detection and clarity verification, this system improves real-time accuracy and reduces false acceptances in comparison to existing systems. By granting access to certified individuals only, ATM crimes can be reduced, and safe, contactless transactions can be performed. Our intelligent ATM access solution represents a new standard in financial security by combining biometric security and IoT technologies and is designed to protect users and to satisfy public health requirements. It offers a smart and very effective solution to combat theft and illegal access, substantially reducing ATM security risks.