Robust Biometric Identification in Networked Society (ROBINS)

• Research unit: COST
• Project number: C01.0096

Key words (English)

biometric identification; automatic speaker recognition; statistical modelling; Internet; missing feature theory; multi-modal processing

Research programs (English)

COST-Action 275 - Biometrics-based recognition of people over the internet

Short description (English)

See abstract

Further information (English)

Full name of research-institution/enterprise: EPF Lausanne STI-IEL-LIDIAP Institut de génie électrique et électronique

Partners and International Organizations (English)

BE, CH, ES, FR, GR, HR, IE, IT, PL, PT, SE, SI, SK, UK

Abstract (English)

The aim of this project is to develop a formulation of the basic concepts of robust biometric recognition, in particular speaker recognition, over wired and wireless Internet data networks in the context of missing data theory, multi-modal feature extraction and statistical modelling, and to make such systems computationally efficient. We investigated methods for detection and modelling unreliable and missing biometric features depending on: 1) background noise in speaker verification, 2) illumination in face verification, 3) inconsistency in dynamic handwritten signature verification, 4) incomplete capture in fingerprint verification. In particular, we developed new techniques for combining pitch and reliable spectral envelope features for robust speaker recognition. Major work has been done in the domain of methodological and technological support for designing and prototyping tools for forensic text-independent speaker recognition, corresponding to challenges of the criminalistic real world and modern communication networks.

References in databases (English)

Swiss Database: COST-DB of the State Secretariat for Education and Research Hallwylstrasse 4 CH-3003 Berne, Switzerland Tel. [phone] Swiss Project-Number: C01.0096