Multimodal fusion vulnerability to non-zero effort (spoof) imposters
Authors:
- 1 December 2010
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abstract
In biometric systems, the threat of “spoofing”, where an imposter will fake a biometric trait, has lead to the increased use of multimodal biometric systems. It is assumed that an imposter must spoof all modalities in the system to be accepted. This paper looks at the cases where some but not all modalities are spoofed. The contribution of this paper is to outline a method for assessment of multimodal systems and underlying fusion algorithms. The framework for this method is described and experiments are conducted on a multimodal database of face, iris, and fingerprint match scores.Keywords
This publication has 10 references indexed in Scilit:
- Robustness of multimodal biometric fusion methods against spoof attacksJournal of Visual Languages & Computing, 2009
- Security and Liveness, Overview2009
- Quality-based Score Level Fusion in Multibiometric Systems2006
- Score normalization in multimodal biometric systemsPattern Recognition, 2005
- Discriminative multimodal biometric authentication based on quality measuresPattern Recognition, 2005
- Incorporating Image Quality in Multi-algorithm Fingerprint Verification2005
- An Introduction to Biometric RecognitionIEEE Transactions on Circuits and Systems for Video Technology, 2004
- Fusion strategies in multimodal biometric verification2003
- Information Fusion in Biometrics2001
- Combining classifiers: A theoretical frameworkPattern Analysis and Applications, 1998