Multi-Biometric Cryptosystem using the Features of Fingerprint and Palmprint for Securing Biometric Templates

Securing the privacy of biometric data has become a critical issue. This paper gives a multi biometrics cryptosystem that combines features of fingerprint and palmprint to overcome several drawbacks of uni biometrics cryptosystem. Preprocessing is a soul of an image processing in which feature enhancement is first step, in this system feature enhancement of input image that is fingerprint and palmprint are performed by applying a series of preprocessing techniques. Gaussian filter is used to independently extract a fingerprint feature which provides accuracy. Then we have extracted the features of both fingerprint and palmprint, in the later step the resultant images are combined in feature level fusion method using fuzzy commitment. After that codeword is generated with the help of Hamming Distance codeword generation method and then hashing is done with the help of MD5 algorithm which generates key. This key is stored and prior matching of the templates is done. We conclude that proposed methodology has better performance as compared to uni biometric cryptosystem approaches using individually only a fingerprint or a palmprint. The multiple biometrics helps to reduce the system error rate.