In this chapter, we have explored presentation attacks in ocular biometric system on smartphones in the visible spectrum. We have discussed both kind of attacks - print attacks and electronic screen attacks. We have employed two publicly available databases that correspond to large-scale image-based artefact, MobiLive 2014 dataset, and for video-based artefacts, PAVID dataset.

Chapter Contents:

• 13.1 Introduction
• 13.2 Database
• 13.2.1 MobILive 2014 Database
• 13.2.2 PAVID Database
• 13.2.2.1 PAVID - Bona Fide iris video database
• 13.2.2.2 PAVID - attack iris video database
• 13.3 Vulnerabilities towards presentation attacks
• 13.3.1 Vulnerability analysis using the PAVID
• 13.4 PAD techniques
• 13.4.1 Metrics for PAD algorithms
• 13.4.2 Texture features for PAD
• 13.4.2.1 Image quality features
• 13.4.2.2 Texture features from LBP
• 13.4.2.3 Texture features from binarized statistical image features
• 13.4.2.4 Laplacian pyramid frequency response features
• 13.5 Experiments and results
• 13.5.1 Results on MobiLive 2014 database
• 13.5.2 Results on the PAVID database
• 13.6 Discussions and conclusion
• Acknowledgments
• References

Inspec keywords: iris recognition; video signal processing; smart phones

Other keywords: electronic screen attacks; presentation attacks; image-based artefact; smartphones; MobiLive 2014 dataset; print attacks; PAVID dataset; video-based artefacts; periocular region

Subjects: Mobile, ubiquitous and pervasive computing; Image recognition; Video signal processing; Computer vision and image processing techniques