Metadata filtering for user-friendly centralized biometric authentication

Abstract

While biometric authentication for commercial use so far mainly has been used for local device unlock use cases, there are great opportunities for using it also for central authentication such as for remote login. However, many current biometric sensors like for instance mobile fingerprint sensors have too large false acceptance rate (FAR) not allowing them, for security reasons, to be used in larger user group for central identification purposes. A straightforward way to avoid this FAR problem is to either request a user unique identifier such as a device identifier or require the user to enter a unique user ID prior to making the biometric matching. Usage of a device identifier does not work when a user desires to authenticate on a previously unused device of a generic type. Furthermore, requiring the user at each login occasion to enter a unique user ID, is not at all user-friendly. To avoid this problem, we in this paper investigate an alternative, most user-friendly approach, for identification in combination with biometric-based authentication using metadata filtering. An evaluation of the adopted approach is carried out using realistic simulations of the Swedish population to assess the feasibility of the proposed system. The results show that metadata filtering in combination with traditional biometric-based matching is indeed a powerful tool for providing reliable, and user-friendly, central authentication services for large user groups.

Signal‐level fusion for indexing and retrieval of facial biometric data

“…A simple and common method for database pre-filtering is the utilisation of geographic and/or demographic metadata to narrow down the potential search space (see e.g. Gehrmann et al [27]); soft biometrics (see e.g. Dantcheva et al [28]) can also be used in an analogous manner.…”

Beneficios de las tecnologías biométricas para la autenticación de usuarios: una revisión sistemática

“…(Zhou & Ren, 2018), (Altuwayjiri et al, 2022), (Albalawi et al, 2022), (El-El-Sofany, 2022), (El-Gazar et al, 2022), (Bharath & Radhakrishna Rao, 2022), (Zhang et al, 2022), (Park & Kim, 2022), (Jaswal et al, 2021), (Laka et al, 2021), (Chin et al, 2021), (Parkinson et al, 2021), (Zhu & Cao, 2021), (Kovalan et al, 2021), (Kausar, 2020), (Wu et al, 2018) 16 2 Ahorro de esfuerzo y tiempo al no tener que recordar una contraseña. (Kao & Chueh, 2023), (Albalawi et al, 2022), (El-El-Sofany, 2022), (Zhang et al, 2022), (Laka et al, 2021), (Parkinson et al, 2021), (Gehrmann et al, 2019), (Lee et al, 2019), (Arun & Sudhakar, 2019) 9 3…”

“…La biometría se encarga de estudiar y analizar las características físicas o el comportamiento propio de cada ser humano, con el objetivo de demostrar la identidad de manera irrefutable, usando aquello que hace a uno diferente. En este sentido, se clasifican en dos categorías muy importantes que son las tecnologías biométricas-fisiológicas y del comportamiento (Gehrmann et al, 2019).…”

Feature Fusion Methods for Indexing and Retrieval of Biometric Data: Application to Face Recognition with Privacy Protection

“…Such methods depend on categorical or weakly discriminative features (e.g. geographic and/or demographic metadata [48] or soft biometrics [49]), whereby the potential search space can be narrowed down quickly prior to considering the actual highly discriminative, but more computationally expensive to compute, biometric features. Conceptually similar two or multi-stage methods operating on weakly discriminative, compact representations (e.g.…”