RUS  ENG
Full version
JOURNALS // Prikladnaya Diskretnaya Matematika. Supplement // Archive

Prikl. Diskr. Mat. Suppl., 2017 Issue 10, Pages 73–76 (Mi pdma324)

Mathematical Methods of Cryptography

About handwritten signature verification

A. V. Epishkina, A. V. Beresneva, S. S. Babkin, A. S. Kurnev, V. U. Lermontov

National Engineering Physics Institute "MEPhI", Moscow

Abstract: Some methods for the online verification of handwritten signatures are presented. The methods are based on the KNN algorithm, a Range Classifier algorithm, a hidden Markov model, and the simplest perceptron neural network. The features of these methods are studied in the course of their implementing and testing with the purpose of the further modifications and the development of the most effective ones for all parameters of the verification procedure.

Keywords: verification of handwritten signature, hidden Markov model, neural network.

UDC: 004.056

DOI: 10.17223/2226308X/10/31



© Steklov Math. Inst. of RAS, 2024