Abstract
Biometrics could accurately identify or verify individuals based upon each person’s unique physical or behavioral characteristics. Among all the biometric traits, fingerprint-based identification is the oldest method which has been successfully used in numerous applications. In recent years, automated fingerprint identification systems (AFIS) have been widely used, such as homeland security, law enforcement, criminal identification. In this thesis, we implement a minutiae-based AFIS, our system tests on 2 databases of fingerprint images, such as Rindex28, Lindex101 from PRIP Lab at NTHU. The system environment is implemented on a system of Intel Core i5 with 3.10 GHz CPU and 8 GB SDRAM running Windows 7. According to the type of top 3 matches, the recognition rates are 99.11% (111/112), 87.87% (355/404), respectively.