Combining Hardwaremetry and Biometry for Human Authentication via Smartphones

Abstract

Abstract. The role of smartphones in our life is ever-increasing. They are used to store and share sensitive data and to perform security critical operation online e.g. home banking transaction or shopping. This leads to the need for a more secure authentication process via mobile phones. Biometrics could be the solution but biometric authentication systems via mobile devices presented so far still do not provide a good trade-off between ease of use and high security level. In this paper we analyze the combination of sensor recognition (hardwaremetry) and iris recognition (biometry) in order to provide a double check of user's identity in one shot, i.e. a single photo of the eye captured by the Smartphone, without the need of additional or dedicated sensors. To the best of our knowledge, this is the first attempt to combine these two aspects.

One-shot Representational Learning for Joint Biometric and Device Authentication

“…In [7], the authors used partial face images acquired using smartphones and employed a weighted sum fusion rule at the score level to combine sensor and biometric recognition. Later, they extended their work to include feature level fusion in [3] and concluded that score level fusion performed comparatively better.…”