Title: Artificial finger detection by spectrum analysis

Authors: Shoude Chang, Jeff Secker, Qinghan Xiao, Brittany Reid, Alexander Bergeron, Wahab Almuhtadi

Addresses: National Research Council of Canada, 1200 Montreal Road, Ottawa, Ontario, K1A 0R6 Canada. ' Defence Research and Development Canada – Ottawa, 3701 Carling Avenue, Ottawa, Ontario, K1A 0Z4 Canada. ' Defence Research and Development Canada – Ottawa, 3701 Carling Avenue, Ottawa, Ontario, K1A 0Z4 Canada. ' Algonquin College, 1385 Woodroffe Avenue, Ottawa, Ontario, K2G 1V8 Canada. ' Algonquin College, 1385 Woodroffe Avenue, Ottawa, Ontario, K2G 1V8 Canada. ' Algonquin College, 1385 Woodroffe Avenue, Ottawa, Ontario, K2G 1V8 Canada

Abstract: To prevent fingerprint spoofing attack, a simple, effective and low-cost approach for distinguishing artificial fingers from real ones based on spectrum analysis is presented in this paper. We measured the spectral features of both artificial and real fingers. The results show that near infrared images of fingers are sufficient to distinguish real fingers from artificial ones, while near ultraviolet images can be used to detect the surface details of fingers. We present a practical and simple algorithm based on these features that can successfully distinguish between real and artificial fingers regardless of the race of the test subject.

Keywords: fingerprint recognition; artificial fingers; anti-spoofing; spectrum analysis; fingerprint spoofing attacks.

DOI: 10.1504/IJBM.2011.042818

International Journal of Biometrics, 2011 Vol.3 No.4, pp.376 - 389

Published online: 24 Jan 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article