Real-time fingerprint identification

Abstract

The pattern matching for fingerprints requires a large amount of data and computation time. Practical fingerprint identification systems require minimal errors and ultrafast processing time to perform real-time verification and identification. By utilizing the two-dimensional processing capability, ultrafast processing speed and non-interfering communication of optical processing techniques, fingerprint identification systems can be implemented in real-time. Among the various pattern matching systems, the joint transform correlator (JTC) has been found to be inherently suitable for real-time matching applications. Among the various JTCs, the fringe-adjusted JTC has been found to yield significantly better correlation output compared to alternate JTCs. In this paper, a fingerprint identification system has been developed based on the fringe-adjusted JTC. Since all pattern matching systems suffer from high sensitivity to distortions, the synthetic discriminant function concept has been incorporated in fringe-adjusted JTC to ensure distortion-invariant fingerprint identification.