Welcome to the InfoSci Platform
Reference Hub8
Fingerprint Liveness Detection Based on Fake Finger Characteristics

Fingerprint Liveness Detection Based on Fake Finger Characteristics

Gian Luca Marcialis, Pietro Coli, Fabio Roli
Copyright: © 2012 |Volume: 4 |Issue: 3 |Pages: 19
ISSN: 1941-6210|EISSN: 1941-6229|EISBN13: 9781466611610|DOI: 10.4018/jdcf.2012070101
Cite Article Cite Article

MLA

Marcialis, Gian Luca, et al. "Fingerprint Liveness Detection Based on Fake Finger Characteristics." IJDCF vol.4, no.3 2012: pp.1-19. http://doi.org/10.4018/jdcf.2012070101

APA

Marcialis, G. L., Coli, P., & Roli, F. (2012). Fingerprint Liveness Detection Based on Fake Finger Characteristics. International Journal of Digital Crime and Forensics (IJDCF), 4(3), 1-19. http://doi.org/10.4018/jdcf.2012070101

Chicago

Marcialis, Gian Luca, Pietro Coli, and Fabio Roli. "Fingerprint Liveness Detection Based on Fake Finger Characteristics," International Journal of Digital Crime and Forensics (IJDCF) 4, no.3: 1-19. http://doi.org/10.4018/jdcf.2012070101

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The vitality detection of fingerprints is currently acknowledged as a serious issue for personal identity verification systems. This problem, raised some years ago, is related to the fact that the 3d shape pattern of a fingerprint can be reproduced using artificial materials. An image quite similar to that of true, alive, fingerprint, is derived if such “fake fingers” are submitted to an electronic scanner. Since introducing hardware dedicated to liveness detection in scanners is expensive, software-based solutions, based on image processing algorithms, have been proposed as alternative. So far, proposed approaches are based on features exploiting characteristics of a live finger (e.g., finger perspiration). Such features can be named live-based, or vitality-based features. In this paper, the authors propose and motivate the use of a novel kind of features exploiting characteristics noticed in the reproduction of fake fingers, that they named fake-based features. Then the authors propose a possibile implementation of this kind of features based on the power spectrum of the fingerprint image. The proposal is compared and integrated with several live-based features at the state-of-the-art, and shows very good liveness detection performances. Experiments are carried out on a data set much larger than commonly adopted ones, containing images from three different optical sensors.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.