Cross-sensor iris spoofing detection using orthogonal features

Iris recognition systems face issues related to spoofing attacks that obfuscate the natural iris texture pattern. This makes it a challenging problem to be used for large-scale high security applications. The paper introduces a rotation-invariant feature-set comprising of Zernike moments and Polar h...

Full description

Saved in:
Bibliographic Details
Published in:Computers & electrical engineering 2019-01, Vol.73, p.279-288
Main Authors: Kaur, Bineet, Singh, Sukhwinder, Kumar, Jagdish
Format: Article
Language:English
Subjects:
Biometric recognition systems
Contact lens
Contact lenses
Cybersecurity
Face recognition
Feature extraction
Feature recognition
Iris spoofing
Polar harmonic transforms
Print attacks
Spoofing
Zernike moments
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Iris recognition systems face issues related to spoofing attacks that obfuscate the natural iris texture pattern. This makes it a challenging problem to be used for large-scale high security applications. The paper introduces a rotation-invariant feature-set comprising of Zernike moments and Polar harmonic transforms that extract local intensity variations for detection of iris spoofing attacks. Experimental results have been implemented on four publicly available iris spoofing databases: IIITD Contact Lens, IIITD Iris Spoofing, Clarkson LivDet-Iris 2015 and Warsaw LivDet-Iris 2015 that include both contact lens and print attack spoofing samples. Experiments demonstrate that the proposed system easily detects spoofing attacks even when acquired using multiple sensors.
ISSN:0045-7906
1879-0755
DOI:10.1016/j.compeleceng.2018.12.002