ONLINE FINGERPRINT VERIFICATION

Abstract

Present day need of our society is to have an identification or verification system, which is fast and reliable. We have various biometrics based verification systems, but fingerprint verification system is one of the fast and reliable systems. In fingerprint verification system minutiae-based representation is most widely used for identification / verification. But due to the limited amount of information present in the minutiae-based representation, it is desirable to explore alternative representations of fingerprints for verification or identification. Various methods of biometric verifications are discussed here. Fingerprint biometrics has been illustrated in detail. In this dissertation a method based on filter bank representation for the verification of fingerprints has been developed. Overall performance of the system is tested on fingerprint database of 50 students named as IITRDB. Results obtained are FAR (False Acceptance Rate) of 23.65% and FRR (False Rejection Rate) of 27.67%. These are good enough for several applications. User friendly software has been developed and implemented for live demonstration and testing.