Using Morphism as a Classifier for Offline Signature Verification

  • Lucindo de Melo UFPE
  • Cleber Zanchettin UFPE
  • Anderson Lira Queiroz UFPE
  • Gliner Dias Alencar UFPE
DOI: https://doi.org/10.5753/sbsi.2014.6156

Abstract


This paper provides contributions to the problem of Off-line signature verification by creating a classifier based on morphism using a technique of triangulation of points. This technique allows checking the distance between the signatures of a particular class, aiming to discrimine false from genuine signatures. In this paper, five scenarios were evaluated using genuine, simulated and random fakes signatures. The fifth scenario use a variable amount of points per experiment, which one obtained the best result, reaching a Global error rate of 0.18.

Published
2014-05-27
How to Cite

Select a Format
DE MELO, Lucindo; ZANCHETTIN, Cleber; QUEIROZ, Anderson Lira; ALENCAR, Gliner Dias. Using Morphism as a Classifier for Offline Signature Verification. In: BRAZILIAN SYMPOSIUM ON INFORMATION SYSTEMS (SBSI), 10. , 2014, Londrina. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2014 . p. 687-698. DOI: https://doi.org/10.5753/sbsi.2014.6156.

Most read articles by the same author(s)