%0 Journal Article
	%A Sanjay Kanade and  Danielle Camara and  Dijana Petrovska-Delacretaz and  Bernadette Dorizzi
	%D 2009
	%J International Journal of Computer and Information Engineering
	%B World Academy of Science, Engineering and Technology
	%I Open Science Index 27, 2009
	%T Application of Biometrics to Obtain High Entropy Cryptographic Keys
	%U https://publications.waset.org/pdf/2115
	%V 27
	%X In this paper, a two factor scheme is proposed to
generate cryptographic keys directly from biometric data, which
unlike passwords, are strongly bound to the user. Hash value of the
reference iris code is used as a cryptographic key and its length
depends only on the hash function, being independent of any other
parameter. The entropy of such keys is 94 bits, which is much higher
than any other comparable system. The most important and distinct
feature of this scheme is that it regenerates the reference iris code by
providing a genuine iris sample and the correct user password. Since
iris codes obtained from two images of the same eye are not exactly
the same, error correcting codes (Hadamard code and Reed-Solomon
code) are used to deal with the variability. The scheme proposed here
can be used to provide keys for a cryptographic system and/or for
user authentication. The performance of this system is evaluated on
two publicly available databases for iris biometrics namely CBS and
ICE databases. The operating point of the system (values of False
Acceptance Rate (FAR) and False Rejection Rate (FRR)) can be set
by properly selecting the error correction capacity (ts) of the Reed-
Solomon codes, e.g., on the ICE database, at ts = 15, FAR is 0.096%
and FRR is 0.76%.
	%P 555 - 559