Secure Fingerprint Authentication in ATM Transaction using Blockchain Technology

Security of ATMs relies on a variety of authentication methods to verify user identity or ensure the integrity of transactions. The traditional PIN-based authentication faces mostly threats that jeopardize its very foundation, including card skimming, pin thefts, and software breaches, amongst others. The paper proposes a fingerprint authentication system for ATMs with integrated blockchain technology. This system employs tree-based classification algorithms to determine fingerprint features, such as ridge patterns and minutiae points, thus enabling accurate and reliable authentication of users. The ensemble classifiers profited boost in classification accuracy and reduced overfitting, thus increasing the robustness of the developed system. The successful authentication of fingerprints has given access to safer functionalities at ATMs-adverse to the fraud risks prevalent for all other methods. With the implementation of blockchain technology, it assures integrity of transactions data, along with transparency and protection from alteration owing to the immutable ledger of transactions. The accountability brought about by the decentralized nature of blockchain makes ATM transactions secured and reliable. This work intends to deliver a prompt, secured, and user-friendly authentication mechanism replacing PIN-based systems while making seamless financial transactions with higher fraud protection possible.