Feature Extraction for Handwritten Signature Verification

Proceedings of The 2nd International Conference on Research in Science, Engineering and Technology

Year: 2019

DOI: https://www.doi.org/10.33422/2nd.icarste.2019.12.836

[Fulltex PDF]

Feature Extraction for Handwritten Signature Verification

Beresneva Anastasia and Epishkina Anna



Handwritten signature verification is one of the promising methods of identity verification, as each person’s signature is unique, especially if its dynamic characteristics are taken into account. The study identified more than 20 characteristics of handwritten signature. The direction of further work is devoted to the definition of significant features for the construction of an effective verification algorithm. To determine the informative features of the signature and use them in building a model of handwritten signature, it is necessary to analyze their statistical characteristics and determine how they are indicative and at the same time stable for one person.

Keywords: Handwritten signature, verification, neural network, mobile application.