University of Salzburg (Peter Wild, Andreas Uhl)


Single-sensor Biometric Surveillance Combining Iris and Face (BioSurveillance) is a FIT-IT Trust in IT-Systems project (no. 819382) on applying a multibiometric combination of iris and face recognition technology in unconstrained environments. The development of new iris recognition algorithms, especially for unconstrained environments, has been an active research topic with a broad range of civilian (e.g. authenticated financial transactions, covert identification in Austrian ATMs, of which up to 50% are already equipped with surveillance cameras) and governmental applications (e.g. insurance cards or passports enhanced with biometric identifiers). While iris-based systems are among the most accurate solutions supporting large-scale identification, a significant drawback of available implementations is the requirement of cooperative acquisition. With the assumed application scenario of semi-constrained capture (the user is assumed to approach the camera such that both irides are at least partially visible) novel approaches towards unconstrained biometric recognition using multiple iris and face recognition algorithms on single-sensor input data aiming to provide competitive results even for visual images with significant quality degradation have been examined. Not only the framework of image enhancement and combination techniques but also custom iris recognition algorithms as by-products of the assessment have been major assets of this project.

Biosurveillance setup


Last modified: 07 November 2015 04:57:29
© Peter Wild. All rights reserved.