<abstract xmlns="http://www.w3.org/1999/xhtml"><p>A combined method for on-line signature verification is presented in this paper. Moreover, all the necessary steps in developing a signature recognition system are described: signature data pre-processing, feature extraction and selection, verification and system evaluation. NNs are used for verification. The influence of the signature forgery type (random and skilled) over the verification results is investigated as well. The experiments are carried out on SUsig database which consists of genuine and forgery signatures of 89 users. The average accuracy is 98.46%.</p></abstract>

A combined method for on-line signature verification is presented in this paper. Moreover, all the necessary steps in developing a signature recognition system are described: signature data pre-processing, feature extraction and selection, verification and system evaluation. NNs are used for verification. The influence of the signature forgery type (random and skilled) over the verification results is investigated as well. The experiments are carried out on SUsig database which consists of genuine and forgery signatures of 89 users. The average accuracy is 98.46%.

A Combined Method for On-Line Signature Verification

Institute of Information and Communication Technologies, 1113 Sofia

Cybernetics and Information Technologies

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

{"article-title":"A Combined Method for On-Line Signature Verification"}

A combined method for on-line signature verification is presented in this paper. Moreover, all the necessary steps in developing a signature recognition system...

A Combined Method for On-Line Signature Verification

Published Online: Jul 15, 2014

Page range: 92 - 97

DOI: https://doi.org/10.2478/cait-2014-0022

Keywords
On-line signature verification, neural networks, feature selection, SUsig database, forgery signatures

© 2014 Desislava Boyadzhieva and Georgi Gluhchev

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

A Combined Method for On-Line Signature Verification

Published Online: Jul 15, 2014

Page range: 92 - 97

DOI: https://doi.org/10.2478/cait-2014-0022

KeywordsOn-line signature verification, neural networks, feature selection, SUsig database, forgery signatures

© 2014 Desislava Boyadzhieva and Georgi Gluhchev

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Keywords
On-line signature verification, neural networks, feature selection, SUsig database, forgery signatures