Hill-climbing attack based on the uphill simplex algorithm and its application to signature verification

  • Authors:
  • Marta Gomez-Barrero;Javier Galbally;Julian Fierrez;Javier Ortega-Garcia

  • Affiliations:
  • Biometric Recognition Group-ATVS, EPS, Universidad Autonoma de Madrid, Madrid, Spain;Biometric Recognition Group-ATVS, EPS, Universidad Autonoma de Madrid, Madrid, Spain;Biometric Recognition Group-ATVS, EPS, Universidad Autonoma de Madrid, Madrid, Spain;Biometric Recognition Group-ATVS, EPS, Universidad Autonoma de Madrid, Madrid, Spain

  • Venue:
  • BioID'11 Proceedings of the COST 2101 European conference on Biometrics and ID management
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

A general hill-climbing attack to biometric systems based on a modification of the downhill simplex algorithm is presented. The scores provided by the matcher are used in this approach to adapt iteratively an initial estimate of the attacked template to the specificities of the client being attacked. The proposed attack is evaluated on a competitive feature-based signature verification system over both the MCYT and the BiosecurID databases (comprising 330 and 400 users, respectively). The results show a very high efficiency of the hill-climbing algorithm, which successfully bypassed the system for over 90% of the attacks with a remarkably low number of scores needed.