Abstract
AbstractForgery of iris patterns to obfuscate an iris recognition system has been a major obstacle in the widescale deployment of iris biometrics. In general, this is termed as an iris presentation attack on the system, which has now become a key challenge in mobile devices as well. The iris scanner in mobile devices is not intelligent to discriminate between real iris patterns and presentation attacks. Therefore, a presentation attack detection mechanism needs to be deployed along with the scanner. This paper proposes a learning-based PAD method to identify the attack patterns by observing their deviation from the real iris. There are two feature extraction methods employed to constitute features individually, from the raw and normalized iris images. Both types of features are combined through feature-level fusion (or concatenation) and an SVM classifier is incorporated to learn the feature representation from those features. The proposed method is evaluated across various attacks, iris scanners, and datasets using the well-known LivDet-2017 dataset to ensure the efficiency and robustness of the approach.KeywordsDomain-specific filtersFeature concatenationFeature dimensionality reductionIris PADVGG-19
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.