Home // eKNOW 2013, The Fifth International Conference on Information, Process, and Knowledge Management // View article

Shape Feature Extraction for On-line Signature Evaluation

Authors:
Jungpil Shin
Weichen Lin

Keywords: shape feature; signature verification; human–computer interaction

Abstract:
In the past few years, banks and companies have increased security by switching from simple static passwords to more dynamic security measures that offer greater protection for users of mobile and web commerce. The most personal method for authentication is analysis of handwritten signatures. Signature evaluation determines whether an individual’s signature is considered “good” or “bad.” A good signature is more complex and difficult to impersonate, whereas a bad signature is simple and easy to impersonate. A signature typically contains many angles, whether big or small. A signature with a higher number of angles is more complex and is considered “good”; therefore, the number of internal angles was calculated to determine the quality of the signature. In this paper, geometry was used to decide the kinds of angles to analyze. After the evaluation, verification was performed to analyze the EER (Equal Error Rate), and it has been concluded that it is best to use the Neighbor method to create the best signature evaluation system.

Pages: 20 to 25

Copyright: Copyright (c) IARIA, 2013

Publication date: February 24, 2013

Published in: conference

ISSN: 2308-4375

ISBN: 978-1-61208-254-7

Location: Nice, France

Dates: from February 24, 2013 to March 1, 2013