Abstract
Apart from ensuring high recognition accuracy, one of the main challenges associated with mobile iris recognition is reliable Presentation Attack Detection (PAD). This paper proposes a method of detecting presentation attacks when the iris image is collected in visible light using mobile devices. We extended the existing database of 909 bona-fide iris images acquired with a mobile phone by collecting additional 900 images of irises presented on a color screen. We explore different image channels in both RGB and HSV color spaces, deep learning-based and geometric model-based image segmentation, and use Local Binary Patterns (LBP) along with the selected statistical images features classified by the Support Vector Machine to propose an iris PAD algorithm suitable for mobile iris recognition setups. We found that the red channel in the RGB color space offers the best-quality input samples from the PAD point of view. In subject-disjoint experiments, this method was able to detect 99.78% of screen presentations, and did not reject any live sample.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: 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.
