A New Multi-spectral Iris Acquisition Sensor for Biometric Verification and Presentation Attack Detection

Multi-spectral iris recognition has increasingly gained interest in recent years. This paper presents the design and implementation of a multi-spectral iris sensor that captures iris images in five different spectral bands. The proposed multi-spectral iris sensor consists of three main parts: (a) Image capture unit (2) Illumination unit and (3) Control unit, each of which is explained in detail. A new multi spectral iris database is captured using the developed sensor to evaluate the efficacy in two scenarios - same-spectral band and cross-spectral band verification. Further, we also present the experiments on Presentation Attack Detection (PAD) using the proposed multi-spectral iris capture device. Experimental results demonstrate that the newly developed sensor indicates reliable and robust performance for both verification and PAD making it a strong candidate sensor for real-life deployment with integrated PAD.