Facial recognition by a compact parallel optical correlator

Abstract

With the progress of information technology, the need for an accurate personalidentification system based on biological characteristics is increasing the demandfor this type of security technology instead of conventional systems usingID cards or pin numbers. Among other features, the face is the mostfamiliar element and is less subject to psychological resistance. As a simpleand compact recognition system satisfying the required performance, weimplemented a hybrid system based on the optical recognition principle usinga multi-level zone plate as a Fourier-transform lens and we report thepreliminary results of their application to face recognition. In this paper, wepresent the design procedure and fabrication process for an improvedversion of a second-generation compact parallel correlator (named COPaCII), the size of which is 20 × 24 × 43 cm3 and weight 6 kg. As a result, weobtained a low error rate of 0% as the false match rate and 0.3% as the falsenon-match rate, thus the COPaC II significant identification security level issufficiently stable. With the aim of further enhancing the throughput androbustness, we conducted performance tests where the system is used as acomputer log-in device and as a pre-screening device for crime investigation.In both experiments, a high rate of successful recognition, such as 90%recognition and 94% rejection rate for log-in, was obtained. Experiments ontwins to check the disguise recognition, and on the effects of changes inbrightness and arbitrary size of images to test its robustness are also included.