Unconstrained palmprint as a smartphone biometric

Abstract

In this work the use of palmprints as an alternative biometric for smartphones is investigated and its feasibility is evaluated across multiple devices using unconstrained acquisition. A novel multi-device database that covers a wide range of camera systems with data obtained in various conditions was built. This database contains palmprint images from 81 users acquired with 5 smartphones and is publicly available to the research community. Extensive experiments were carried out and their results placed into the context of existing, more constrained palmprint databases acquired with smartphones. The results outline the potential of palmprints to be used as an efficient secondary biometric deployed on consumer devices.