Comparative Analysis of Copy Detection Patterns and Physical Unclonable Functions in Authentication Systems: A Mobile Phone Perspective

In this paper, we perform a comparative analysis of Copy Detection Patterns (CDP) and Physical Unclonable Functions (PUFs) within the context of anti-counterfeiting technologies, examining their effectiveness against machine-learning based attacks and their functioning across different mobile imaging devices. In the evaluation involving both iPhone 12 Pro and Samsung Galaxy S20, we assess the authentication performance of CDP and PUFs integragted in printed labels produced on industrial printer. Our approach includes the creation of machine-learning based counterfeits to challenge the resilience of each authentication method. The outcomes, quantified via Receiver Operating Characteristic analysis, reveal the robustness of CDP and PUFs against counterfeiting, underscoring CDP's consistency across mobile devices and highlighting PUFs's sensitivity to the variations in mobile imaging quality. This research not only provides insights into the operational strengths and vulnerabilities of CDP and PUFs but also proposes recommendations for enhancing the practical applications of these technologies in protection against counterfeit goods.