Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control and Artificial Intelligence
Synthetic Fingerprint-Image Generation
ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 3
Biometric User Authentication for IT Security: From Fundamentals to Handwriting (Advances in Information Security)
Protecting Face Biometric Data on Smartcard with Reed-Solomon Code
CVPRW '06 Proceedings of the 2006 Conference on Computer Vision and Pattern Recognition Workshop
Biometrics based Asymmetric Cryptosystem Design Using Modified Fuzzy Vault Scheme
ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 04
IEEE Transactions on Pattern Analysis and Machine Intelligence
EURASIP Journal on Advances in Signal Processing
Biometric key binding: fuzzy vault based on iris images
ICB'07 Proceedings of the 2007 international conference on Advances in Biometrics
Biometric Hash algorithm for dynamic handwriting embedded on a java card
BioID'11 Proceedings of the COST 2101 European conference on Biometrics and ID management
Two-factor face authentication using matrix permutation transformation and a user password
Information Sciences: an International Journal
Hi-index | 0.00 |
Biometric Hash algorithms, also called BioHash, are commonly designed to ensure template protection to its biometric raw data. They provide a certain level of robustness against input variability to assure reproducibility by compensating for intra-class variation of the biometric raw data. This concept can be a potential vulnerability. In this paper, we present two different approaches which exploit this vulnerability to reconstruct raw data out of a given BioHash for handwriting introduced in [1]. The first method uses manual user interaction combined with a genetic algorithm and the second approach uses a spline interpolation function based on specific features to generate raw data. Although they are hard to compare due to the different design, we do compare them with respect to their ability to reconstruct raw data within specific scenarios (constrained and unconstrained time and trails), time consumption during the reconstruction and usability. To show the differences, we evaluate using 250 raw data sets (10 individuals overall) consisting of 5 different handwriting semantics. We generate 50 BioHash vectors out of the raw data and use these as reference data to generate artificial raw data by using both attack methods. These experimental results show that the interactive approach is able to reconstruct raw data more accurate, compared to the automatic method, but with a much higher reconstruction time. If several hundred raw data samples are generated by the automatic approach the chance rises that one of the samples achieve equal or even better results than the interactive approach.