Advanced Multi-Perspective Enrolment in Finger Vein Recognition

Abstract

Finger vein recognition deals with the recognition of subjects based on their venous pattern within the fingers. It has been shown that its recognition accuracy heavily depends on a good alignment of the acquired samples. There are several approaches that try to reduce the impact of finger misplacement. However, none of these approaches is able to prevent all possible types of finger misplacements. As finger vein scanners are evolving towards contact-less acquisition, alignment problems, especially due to longitudinal finger rotation, are becoming even more important. Along with rotation detection and correction, capturing the vein pattern from multiple perspectives, as e.g. in multiple-perspective enrolment (MPE, [1]), is a way to tackle the problem of longitudinal finger rotation. Involving multiple cameras increases cost and complexity of the capturing devices, and therefore their number should be kept to a minimum. Perspective multiplication for MPE (PM-MPE, [2]) successfully reduces the number of cameras needed during enrolment while keeping the recognition rates at a high level. So far, (PM-)MPE has only been applied using Maximum curvature features (MC, [3]). This work analyses further approaches to improve the their recognition rates and investigates the applicability of (PM-)MPE to recognition schemes using features other than MC.