Study of the robustness of a transformation-based multi-biometric template schemes protection

As thenumber of applications that use multimodal biometrics grows, potentialchallenges to privacy and protection emerge. New multi-biometric technologieshave been suggested in the context of privacy and confidentiality: templateprotection systems. For the relevant defensive technologies, very currentsolutions have been presented. Their decision on the optimal strategy is stillbeing considered. This paper provides an evaluation of multi-biometricprotection systems according to a transformation proposing some measures andmetrics to test its performances and sturdiness faced against attacks. Thesemeasures enable evaluating the success of the approach in various scenarios. Inthis perspective, this article proposes a transformation-based multi-modalbiometric protection system that focuses on BioHashing by merging fingerprintmodalities. The fingerprint images are classified into three instances andrepresented by a feature vector using a 2D Log-Gabor filter. A biohashing is appliedto its vectors to create Biocode vectors. Following that, a multi-instancescore fusion formed by Biocodes of the fingerprints is merged at the matchinglevel. Finally, to get a high score, multi-sample score fusion is used.Furthermore, a transformation-based protection model that focuses on BioHashingdemonstrates how specific security and privacy may be validated.