Evolutionary Methods for Generating Synthetic MasterPrint Templates: Dictionary Attack in Fingerprint Recognition

Recent research has demonstrated the possibility of generating "Masterprints" that can be used by an adversary to launch a dictionary attack against a fingerprint recognition system. Masterprints are fingerprint images that fortuitously match with a large number of other fingerprints thereby compromising the security of a fingerprint-based biometric system, especially those equipped with small-sized fingerprint sensors. This work presents new methods for creating a synthetic MasterPrint dictionary that sequentially maximizes the probability of matching a large number of target fingerprints. Three techniques, namely Covariance Matrix Adaptation Evolution Strategy (CMA-ES), Differential Evolution (DE) and Particle Swarm Optimization (PSO), are explored. Experiments carried out using a commercial fingerprint verification software, and public datasets, show that the proposed approaches performed quite well compared to the previously known MasterPrint generation methods.

1. Introduction

A growing number of mobile devices like laptops, tablets or smartphones are utilizing automated fingerprint verification for user authentication. Often, the sensors embedded in these devices capture only a limited portion of the whole fingerprint and, therefore, multiple such partial fingerprints are acquired and stored for the same finger during enrollment [4]. Further, a user is often permitted to enroll the partial fingerprints corresponding to multiple fingers in an attempt to increase the usability of the system. The user is deemed to be successfully authenticated, if the input print acquired during authentication matches anyone of the stored partial prints obtained during enrollment.