2019/08/27 by Lázaro J. González‐Soler, González-Soler, Lázaro J., Marta Gomez‐Barrero +7
Computer Science · Social Sciences · #Biometric Identification and Security #Computer Vision and Pattern Recognition (cs.CV) #Digital Media Forensic Detection #FOS: Computer and information sciences #Forensic Fingerprint Detection Methods
paper · pdf · doi:10.48550/arxiv.1908.10163
openalex publication_date 2019/08/27 · openalex created_date 2022/07/28 · openalex updated_date 2026/07/28
Fingerprint-based biometric systems have experienced a large development in\nthe last years. Despite their many advantages, they are still vulnerable to\npresentation attacks (PAs). Therefore, the task of determining whether a sample\nstems from a live subject (i.e., bona fide) or from an artificial replica is a\nmandatory issue which has received a lot of attention recently. Nowadays, when\nthe materials for the fabrication of the Presentation Attack Instruments (PAIs)\nhave been used to train the PA Detection (PAD) methods, the PAIs can be\nsuccessfully identified. However, current PAD methods still face difficulties\ndetecting PAIs built from unknown materials or captured using other sensors.\nBased on that fact, we propose a new PAD technique based on three image\nrepresentation approaches combining local and global information of the\nfingerprint. By transforming these representations into a common feature space,\nwe can correctly discriminate bona fide from attack presentations in the\naforementioned scenarios. The experimental evaluation of our proposal over the\nLivDet 2011 to 2015 databases, yielded error rates outperforming the top\nstate-of-the-art results by up to 50 % in the most challenging scenarios. In\naddition, the best configuration achieved the best results in the LivDet 2019\ncompetition (overall accuracy of 96.17 %).\n