Offline signature verification using a single master signature for building a signer’s profile
Previous studies on offline signature verification have assumed that there will be at least three sample signatures available from which a profile of an individual signer can be constructed. This paper outlines three signature comparison measures for use where only one master signature is available,...
Main Authors: | , |
---|---|
Format: | Article |
Published: |
CESER Publications
2017
|
Subjects: |
_version_ | 1796863244144476160 |
---|---|
author | Abulfatah, G. M. Ahma, M. N. |
author_facet | Abulfatah, G. M. Ahma, M. N. |
author_sort | Abulfatah, G. M. |
collection | ePrints |
description | Previous studies on offline signature verification have assumed that there will be at least three sample signatures available from which a profile of an individual signer can be constructed. This paper outlines three signature comparison measures for use where only one master signature is available, along with techniques for constructing a suitable threshold. Using these comparison measures and thresholds devised for seen forgeries, an EER of 15% for seen forgeries and FARs of less than 3% and 1% for unseen and random forgeries respectively has been achieved. |
first_indexed | 2024-03-05T20:23:50Z |
format | Article |
id | utm.eprints-80715 |
institution | Universiti Teknologi Malaysia - ePrints |
last_indexed | 2024-03-05T20:23:50Z |
publishDate | 2017 |
publisher | CESER Publications |
record_format | dspace |
spelling | utm.eprints-807152019-06-27T06:20:40Z http://eprints.utm.my/80715/ Offline signature verification using a single master signature for building a signer’s profile Abulfatah, G. M. Ahma, M. N. QA75 Electronic computers. Computer science Previous studies on offline signature verification have assumed that there will be at least three sample signatures available from which a profile of an individual signer can be constructed. This paper outlines three signature comparison measures for use where only one master signature is available, along with techniques for constructing a suitable threshold. Using these comparison measures and thresholds devised for seen forgeries, an EER of 15% for seen forgeries and FARs of less than 3% and 1% for unseen and random forgeries respectively has been achieved. CESER Publications 2017 Article PeerReviewed Abulfatah, G. M. and Ahma, M. N. (2017) Offline signature verification using a single master signature for building a signer’s profile. International Journal of Imaging and Robotics, 17 (2). pp. 87-94. ISSN 2231-525X |
spellingShingle | QA75 Electronic computers. Computer science Abulfatah, G. M. Ahma, M. N. Offline signature verification using a single master signature for building a signer’s profile |
title | Offline signature verification using a single master signature for building a signer’s profile |
title_full | Offline signature verification using a single master signature for building a signer’s profile |
title_fullStr | Offline signature verification using a single master signature for building a signer’s profile |
title_full_unstemmed | Offline signature verification using a single master signature for building a signer’s profile |
title_short | Offline signature verification using a single master signature for building a signer’s profile |
title_sort | offline signature verification using a single master signature for building a signer s profile |
topic | QA75 Electronic computers. Computer science |
work_keys_str_mv | AT abulfatahgm offlinesignatureverificationusingasinglemastersignatureforbuildingasignersprofile AT ahmamn offlinesignatureverificationusingasinglemastersignatureforbuildingasignersprofile |