A fingerprint verification system is vulnerable to attacks led through the fingertip replica of an enrolled user. The countermeasure is a software/hardware module called fingerprint presentation attacks detector (FPAD) that is able to detect images coming from a real (live) and a spoof (fake) fingertip. We focused our work on the so-called software-based solutions that use a classifier trained with a collection of live and fake fingerprint images in order to determine the liveness level of a finger, that is, the probability that the submitted fingerprint image is not a replica. The chapter goal is to give an overview of FPAD systems by focusing on the problem of the interoperability among different capture devices. In other words, the FPAD performance variation arises when the capture device is substituted by another one, for example, due to upgrading reasons. After a brief summary of the main and most effective state-of-the-art approaches to feature extraction, we introduce the interoperability FPAD problem from the image captured by the fingerprint sensor to the impact on the related feature space and classifier. In particular, we take into account the so-called textural descriptors used for FPAD. We review the state of the art in order to see if and how this problem has been already treated. Finally, a possible solution is suggested and a set of experiments is done to investigate its effectiveness.

Interoperability among capture devices for fingerprint presentation attacks detection

Ghiani, L.
;
Marcialis, G. L.
2019-01-01

Abstract

A fingerprint verification system is vulnerable to attacks led through the fingertip replica of an enrolled user. The countermeasure is a software/hardware module called fingerprint presentation attacks detector (FPAD) that is able to detect images coming from a real (live) and a spoof (fake) fingertip. We focused our work on the so-called software-based solutions that use a classifier trained with a collection of live and fake fingerprint images in order to determine the liveness level of a finger, that is, the probability that the submitted fingerprint image is not a replica. The chapter goal is to give an overview of FPAD systems by focusing on the problem of the interoperability among different capture devices. In other words, the FPAD performance variation arises when the capture device is substituted by another one, for example, due to upgrading reasons. After a brief summary of the main and most effective state-of-the-art approaches to feature extraction, we introduce the interoperability FPAD problem from the image captured by the fingerprint sensor to the impact on the related feature space and classifier. In particular, we take into account the so-called textural descriptors used for FPAD. We review the state of the art in order to see if and how this problem has been already treated. Finally, a possible solution is suggested and a set of experiments is done to investigate its effectiveness.
2019
978-3-319-92626-1
978-3-319-92627-8
Software; Signal Processing; 1707; Artificial Intelligence
File in questo prodotto:
File Dimensione Formato  
B11.pdf

Solo gestori archivio

Tipologia: versione post-print
Dimensione 5.43 MB
Formato Adobe PDF
5.43 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11584/260995
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 3
  • ???jsp.display-item.citation.isi??? ND
social impact