Multibiometric based authentication using feature level fusion

Security is the most important thing in the world. Password is used for security but it does not provide the effective security. So single biometric is preferred but due to lack of reliability, it is not efficient. So Multibiometric is used. Multibiometric provides more reliability. Fingerprint and Iris are the most unique features because it is unchangeable anywhere most of the time, compare to other biometrics. In the proposed approach 5 steps are followed in enrollment phase i) Feature extraction of fingerprint and iris ii) Fusion of extracted features iii) Key is generated from the fused feature which is of greater than 128 bit which is enough for AES encryption iv) AES encryption v) Hash Encoding vi) AES decryption. Message is subjected to hash and hash of message is obtained and it is given to AES encryption as a message and the key is obtained from the fused feature. In the verification phase, decryption is to be done. Message is obtained when the encrypted value from the enrollment phase and verification phase are same. Fingerprint is obtained from publicly available sources and Iris is obtained from CASIA Iris database. From this system performance of False acceptance rate and False rejection rate are highly reduced.