A novel hand reconstruction approach and its application to vulnerability assessment

Abstract

Esta es la versión de autor del artículo publicado en: This is an author produced version of a paper published in:Information Sciences 238 (2014) AbstractThe present work proposes a novel probabilistic method to reconstruct a hand shape image from its template. We analyse the degree of similarity between the reconstructed images and the original samples in order to determine whether the synthetic hands are able to deceive hand recognition systems. This analysis is made through the estimation of the success chances of an attack carried out with the synthetic samples against an independent system. The experimental results show that there is a high chance of breaking a hand recognition system using this approach. Furthermore, since it is a probabilistic method, several synthetic images can be generated from each original sample, which increases the success chances of the attack.

Protected Facial Biometric Templates Based on Local Gabor Patterns and Adaptive Bloom Filters

“…As any other security technology, biometric systems are exposed to external attacks that can compromise their security. In particular, it has been proven that it is feasible to recover the biometric trait (i.e., iris [2], face [3], handshape [4] or fingerprint [5]) from the information contained in the stored biometric reference data (template). This fact poses a severe security and privacy problem: a skilled impostor could have access to our biometric data.…”

Reversing the irreversible: A survey on inverse biometrics

“…In other words, they take advantage of the information conveyed in bona fide biometric templates to gain some knowledge of the underlying biometric information, hence violating the privacy of the owner. Such methodologies have already been applied to fingerprint [6,26,25], iris [28,24], handshape [34], face [40,12], or handwriting [81,82].…”

“…Similarly, Gomez-Barrero et al proposed in [34] a probabilistic inverse biometrics method based on a combination of a handshape images generator and an adaptation of the Nelder-Mead simplex algorithm, which had been previously used to recover face images in [84].…”

“…Recently, this official recognition of the key importance of the irreversibility of templates in biometrics has fostered and strengthened even more the interest that the biometric community has shown on the study of this field over the last 15 years. This interest has led to: 1) new reconstruction algorithms for different biometric characteristics [32,33]; 2) evaluation methodologies to determine the risk posed by these reconstruction approaches [24,34]; and 3) new countermeasures to protect the systems against this potential threat [35]. All these works have constituted a new research area commonly referred to as inverse biometrics.…”

Failure of affine‐based reconstruction attack in regenerating vascular feature points

“…Generally, attacks that exploit the information related to comparison scores can be categorised into either hill-climbing attacks or affine-based reconstruction attacks. In hill-climbing attacks [10][11][12] the attacker only needs to have access to the comparison scores issued by the biometric recognition algorithm. He can start the attack by presenting an arbitrary (possibly synthetic) biometric sample for comparison with a targeted biometric reference (BR) in the system.…”