Biometric Parameter Based Cryptographic Key Generation
Rashi Bais1, K.K.Mehta2
1Rashi Bais, Pursuing M.Tech., Department of Computer Science & Engineering, SSCET-Bhilai (C.G.), India.
2Dr. K.K. Mehta, Professor & Head, Department of Computer Science & Engineering, SSCET-Bhilai (C.G.), India.
Manuscript received on may 27, 2012. | Revised Manuscript received on June 22, 2012. | Manuscript published on June 30, 2012. | PP: 157-160 | Volume-1 Issue-5, June 2012 | Retrieval Number: E0456061512/2012©BEIESP

Open Access | Ethics and  Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: A method is proposed for generation of unique cryptographic key which is generated using biometrics of the user, which are stable throughout person’s lifetime. The proposed approach reduces the cost associated with lost keys, addresses non-repudiation issues and provides increased security of digital content. This approach has reduced the complicated sequence of the operation to generate crypto keys as in the traditional cryptography system. The key is derived directly from the biometric data and is not stored in the database, since it creates more complexity to crack or guess the cryptographic keys. We evaluated our technique using 50 different fingerprint samples, and found that an error-free key can be reproduced reliably with a 99.5% success rate. This approach is implemented in MATLAB and can generate variable size cryptographic key, with minimum amount of time complexity, which is aptly suited for any real time cryptography. 
Keywords: Cryptography, Biometrics, Minutiae points, Morphological Operation, Histogram Equalization, Crossing Number.