On Iris Spoofing Using Print Attack

摘要: Human iris contains rich textural information which serves as the key for biometric identifications. It is very unique and one of most accurate modalities. However, spoofing techniques can be used to obfuscate or impersonate identities increase risk false acceptance rejection. This paper revisits recognition with attacks analyzes their effect on performance. Specifically, print attack contact lens variations mechanism. observed that lens, individually in conjunction, significantly change inter-personal intra-personal distributions thereby possibility deceive systems. The also presents IIITD database, over 4800 images pertaining 100 individuals due sensor, attack. Finally, shows cost effective descriptor approaches may help counter-measuring spooking attacks.