Advancements in ATM Security for Movement and Tampering Detection: A Review

Abstract

Advancements in ATM security have become critical due to rising incidents of theft and vandalism. This review aims to evaluate the current state of movement and tampering detection technologies in ATMs, focusing on the integration of advanced machine learning (ML) and deep learning (DL) techniques. Motivated by the need for robust, real-time security solutions, the review addresses limitations in existing systems, such as inadequate anomaly detection and high false-positive rates. The objective is to synthesize advancements in ML and DL methods, including YOLO-based approaches, to enhance ATM security. By examining various methodologies, this review highlights the strengths and weaknesses of different detection systems and proposes directions for future improvements, particularly through the application of the latest YOLO models.