Iris biometrie: Is the near-infrared spectrum always the best?

Abstract

Previous work on iris recognition focused on either Visible Light (VL), Near-Infrared (NIR) imaging or the fusion between them. However, limited numbers of works have compared the iris biometric performance under both VL and NIR spectrum using images taken from the same subject. In this paper, we explore the differences in iris recognition performance across the VL and NIR spectrum. In addition, we investigate the possibility of cross-channel matching between the VL and NIR imaging. We carried out our experiments on the UTIRIS database which contains iris images taken under both the VL and NIR spectrum for the same subject. This paper is amongst the first studies which compares the performance of iris biometric under the NIR and VL spectrum and produces comparative experiments between these types of data. Experimental results indicate that the VL and NIR images provide complementary features for the iris pattern and their fusion improves the recognition performance. In addition, the experiments indicate that cross-channel matching between VL and NIR images is feasible.