Depth-of-field extension method for iris recognition based on multi-wavelength illumination and dispersion principle

Abstract

This article describes the design of an optical system using multi-wavelength illumination and dispersion principle to achieve iris recognition imaging with a large depth of field in noncooperative environments. Through the use of optical design software, the iris acquisition imaging system was designed with the modulation transfer function (MTF) curve serving as the evaluation criteria for this system. It was found that at different object distances, the MTF value at a spatial frequency of 100 lp/mm is greater than 0.33, indicating that the iris acquisition imaging system has a significant depth of field. To further validate the performance of the system, we set up an actual optical platform based on the parameters designed by the optical design software for the iris acquisition imaging system and conducted iris image acquisition experiments at different object distances, obtaining iris images of three different individuals. We performed algorithm verification on the iris recognition of the three sets of iris images and found that the average interclass matching result was 0.44, while the average intraclass matching result was 0.30, demonstrating a high level of accuracy in iris recognition. This indicates that the iris images obtained by the system have a high resolution, thereby confirming the feasibility of depth extension using multi-wavelength illumination and lens chromatic aberration imaging principle. It is worth mentioning that this method is not only applicable to iris recognition but also holds reference value in other biometric recognition fields.