Abstract

LEGON, Carlos Miguel et al. New probabilistic model on graphical authentication. EAC [online]. 2019, vol.40, n.3, pp. 92-104.  Epub Sep 08, 2019. ISSN 1815-5928.

Authentication is essential in the security of modern digital information processing services. Alphanumeric passwords are the most used, but they have weaknesses that make them vulnerable to various attacks based on probabilistic models. One of the alternatives is graphical authentication. The probabilistic models in graphical authentication are applied to estimate the most likely key to be selected in each image by the user to be registered. In this work a new probabilistic model of graphic authentication is proposed, its main contribution is that it allows quantifying the degree of authenticity of each user. It is confirmed experimentally that the proposed model is effective and allows to measure in practice the authenticity level of authenticated users.

Keywords : graphical authentication; probabilistic model; password.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )