Fuzzy Commitment Scheme for Masked Iris Codes
Osama M. Ouda · International Journal of Computer Applications · 2013
The fuzzy commitment scheme is one of most popular biometric cryptosystems that aim at securing cryptographic keys using biometrics.Because of the high recognition accuracy exhibited by the iris, iris-based fuzzy commitment schemes, among other modalities, provide the most practical performance rates.Unfortunately, existing iris-based fuzzy commitment schemes do not incorporate noise masks, generated along with iris-codes to highlight unwanted regions of the iris, because there is no way to know the mask of the decoding iris sample in advance.Therefore, the decoding accuracy of iris-based fuzzy commitment schemes is much less than the recognition accuracy of the underlying iris recognition system.This paper presents an iris-based fuzzy commitment scheme that uses the noise mask of the encoding iris sample at both encoding and decoding stages.Experimental results show that the proposed scheme provides a remarkable improvement in the decoding accuracy of iris-based fuzzy commitment schemes.