MALICIOUS QR CODE DETECTION: A COMPREHENSIVE REVIEW OF METHODS AND CHALLENGES

Abstract

QR codes have become popular to permit the speedy and touch-free access to online services in the spheres of payments, health care, transport, and smart surroundings. Nevertheless, their opaque and machine-readable properties have made them a very appealing target of attacks by phishing, the distribution of malware, financial fraud, and unauthorized redi-rection, also known as quishing attacks. Conventional blacklist- based and rule-based security systems are mostly inefficient against such threats, especially in dynamic and physical worlds. The current paper is a review of the research published in the field of malicious QR code detection between 2023 and 2025. The systematically reviewed machine learning, deep learning, hybrid detection frameworks, and infrastructure-level security mechanisms are analyzed, such as URL-based analysis, image- level QR inspection, multimodal learning, and preventive mech-anisms, such as blockchain-based verification and digital water- marking. The important areas like feature extraction techniques, data properties, assessment procedures, and reported trends of performance are scrutinized. It also addresses that if a model works well in theory, it must also handle real data, real users, limited resources, and possible attacks in the real world. Judging by the reviewed literature, this paper establishes an open research gap and presents the possible future directions in the creation of powerful, scalable, and user-friendly QR code security solutions acceptable in the real-world implementation.