Biometric recognition, principles and privacy protection
The biometric comparator is the core engine of most biometric recognition systems. Its current realisations are predominantly based on machine learning. In the talk I will first review the underlying principles, which differ from those of standard machine-learning based classification, and their relation to statistical optimality. Next, I will address various security and privacy risks of the storage and processing of biometric data and discuss proposed privacy enhancing solutions.
Raymond Veldhuis graduated from the University of Twente, The Netherlands, in 1981. He received the Ph.D. degree from Nijmegen Univer- sity on a thesis entitled Adaptive Restoration of Lost Samples in Discrete-Time Signals and Digital Images, in 1988. From 1982 to 1992, he was a Researcher with Philips Research Laboratories, Eindhoven, in various areas of digital signal processing. From 1992 to 2001, he was involved in the field of speech processing. He is currently a Full Professor in Biometric Pattern Recognition with the University of Twente. His main research topics are face recognition, fingerprint recognition, vascular pattern recognition, multibiometric fusion, and biometric template protection. He is especially interested in the connections between biometric recognition, machine learning, statistical pattern recognition and information theory. The research is both applied and fundamental.