International Journal of Computer Applications
Fingerprint recognition is one of the research hotspots of biometrics techniques. Fingerprints are the most widely used biometric feature for identification and verification in the field of biometrics. The traditional fingerprint recognition systems have such disadvantages as high computation complexity, low speed, low recognition rate to uncompleted or defiled fingerprints, and not robust. In this paper, the authors propose a multimodel fingerprint identification and verification method based on pattern recognition, which emphasizes global features of fingerprint.