An Analysis of Minutiae Matching Strength
AVBPA '01 Proceedings of the Third International Conference on Audio- and Video-Based Biometric Person Authentication
Biometric Hash based on Statistical Features of Online Signatures
ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 1 - Volume 1
Biometric User Authentication for IT Security: From Fundamentals to Handwriting (Advances in Information Security)
Reverse-engineer methods on a biometric hash algorithm for dynamic handwriting
Proceedings of the 12th ACM workshop on Multimedia and security
CMS'10 Proceedings of the 11th IFIP TC 6/TC 11 international conference on Communications and Multimedia Security
Hi-index | 0.00 |
In some biometric verification systems smart cards are used to store personal biometric data of a person instead of storing them in a database. If smart cards are correctly authenticated such as with cryptographic signatures, attacks to biometric databases are reduced like template replacement or cross-matching of biometric databases. Furthermore, smart cards are used to match reference data with the actual data of a claimed identity (e.g. [1], [2] and [3]); these systems are called matching-on-card (MOC). In this paper we present a system which besides matching, storing and decision of biometric templates also implements the feature extractor on a smart card to increase the security level and therefore minimize attack possibilities. In this work a exemplary Java Card as widely deployed smart card environment in today business applications is used as implementation platform and a biometric hash algorithm for dynamic handwriting introduced in [4] as biometric user authentication method. Our goal is to evaluate the processing time performance and EER to show the overall tendency. Due to the limited hard- and software resources of a Java Card, a feature extractor with reduced features (9 features), selected based on uncomplex implementation and fast determination time, is deployed. First experimental results show that a biometric hash algorithm for dynamic handwriting, embedded on a Java Card including the feature extraction, is capable of biometric user verification. However, processing time measurements of the first experimental non-time-optimized test system show that real-time applications are not suitable. To show the verification performance we use 500 raw data sets. Test results show an average EER of 28.5%, whereas a reference biometric hash algorithm (103 features) executed on a standard computer achieves an average EER of 4.86%. Furthermore we compare the performance with an existing DSP (digital signal processor) implementation.