No abstract

In the last few decades, we have witnessed a large-scale deployment of biometric systems in different life applications replacing the traditional recognition methods such as passwords and tokens. We approached a time where we use biometric systems in our daily life. On a personal scale, the authentication to our electronic devices (smartphones, tablets, laptops, etc.) utilizes biometric characteristics to provide access permission. Moreover, we access our bank accounts, perform various types of payments and transactions using the biometric sensors integrated into our devices. On the other hand, different organizations, companies, and institutions use biometric-based solutions for access control. On the national scale, police authorities and border control measures use biometric recognition devices for individual identification and verification purposes.

Fingerprint recognition systems have proven significant performance in many services such as forensics, border control, and mobile applications. Even though fingerprint systems have shown high accuracy and user acceptance, concerns have raised questions about the possibility of having our fingerprint pattern stolen and presented to the system by an imposter. In this paper, we propose a dynamic presentation attack detection mechanism that seeks to mitigate presentation attacks. The adopted mechanism extracts the variation of global fingerprint features in video acquisition scenario and uses it to distinguish bona fide from attack presentations. For that purpose, a dynamic dataset has been collected from 11 independent subjects, 6 fingerprints per user, using thermal and optical sensors. A total of 792 bona fide presentations and 2772 attack presentations are collected. The final PAD subsystem is evaluated based on the standard ISO/IEC 30107-3. Considering SVM classification and 3 folds cross validation, the obtained error rates at 5% APCER are 18.1% BPCER for the thermal subset and 19.5% BPCER for the optical subset.

No abstract

Biometric recognition is already a big player in how we interact with our phones and access control systems. This is a result of its comfort of use, speed, and security. For the case of border control, it eases the task of person identification and blacklist checking. Although the performance rates for verification and identification have dropped in the last decades, protection against vulnerabilities is still under heavy development. This paper will focus on the detection of presentation attacks in fingerprint biometrics, i.e., attacks that are performed at the sensor level, and from a hardware perspective. Most research on presentation attacks has been carried out on software techniques due to its lower price as, in general, hardware solutions require additional subsystems. For this paper, two low-cost handheld microscopes with special lighting conditions were used to capture real and fake fingerprints, obtaining a total of 7704 images from 17 subjects. After several analyses of wavelengths and classification, it was concluded that only one of the wavelengths is already enough to obtain a very low error rate compared with other solutions: an attack presentation classification error rate of 1.78% and a bona fide presentation classification error rate (BPCER) of 1.33%, even including non-conformant fingerprints in the database. On a specific wavelength, a BPCER of 0% was achieved (having 1926 samples). Thus, the solution can be low cost and efficient. The evaluation and reporting were done following ISO/IEC 30107-3. INDEX TERMS Biometrics, fingerprint biometrics, presentation attack detection.