An offline writer independent signature verification method with robustness against scalings and rotations
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Tipo de documento: | Dissertação |
Idioma: | por |
Título da fonte: | Biblioteca Digital de Teses e Dissertações da UFRGS |
Texto Completo: | http://hdl.handle.net/10183/247543 |
Resumo: | Handwritten signatures are still one of the most used and accepted methods for user au thentication. They are used in a wide range of human daily tasks, including applications from banking to legal processes. The signature verification problem consists of verifying whether a given handwritten signature was generated by a particular person, by com paring it (directly or indirectly) to genuine signatures from that person. In this research work, a new offline writer-independent signature verification method is introduced (named VerSig-R), based on a combination of handcrafted Moving Least-Squares features and features transferred from a convolutional neural network. In our experiments, VerSig-R outperforms state-of-the-art techniques on Western-style signatures (CEDAR dataset), while also obtaining competitive results on South Asian-style handwriting (Bangla and Hindi datasets). Furthermore, a wide range of experiments demonstrate that VerSig-R is the most robust in relation to differences in scale and rotation of the signature images. This work also presents a discussion on dataset bias and on cross-dataset performance of VerSig-R, as well as a small user study showing that the proposed technique outperforms the expected human accuracy on the signature-verification task. Finally, a discussion on the impact of the number of signature examples (per writer) used during training on performance and execution time is presented. |
id |
URGS_0d66afae215d3efa405f9df99e72ddfb |
---|---|
oai_identifier_str |
oai:www.lume.ufrgs.br:10183/247543 |
network_acronym_str |
URGS |
network_name_str |
Biblioteca Digital de Teses e Dissertações da UFRGS |
repository_id_str |
1853 |
spelling |
Pachas, Felix Eduardo HuarotoGastal, Eduardo Simões Lopes2022-08-20T04:55:47Z2022http://hdl.handle.net/10183/247543001146961Handwritten signatures are still one of the most used and accepted methods for user au thentication. They are used in a wide range of human daily tasks, including applications from banking to legal processes. The signature verification problem consists of verifying whether a given handwritten signature was generated by a particular person, by com paring it (directly or indirectly) to genuine signatures from that person. In this research work, a new offline writer-independent signature verification method is introduced (named VerSig-R), based on a combination of handcrafted Moving Least-Squares features and features transferred from a convolutional neural network. In our experiments, VerSig-R outperforms state-of-the-art techniques on Western-style signatures (CEDAR dataset), while also obtaining competitive results on South Asian-style handwriting (Bangla and Hindi datasets). Furthermore, a wide range of experiments demonstrate that VerSig-R is the most robust in relation to differences in scale and rotation of the signature images. This work also presents a discussion on dataset bias and on cross-dataset performance of VerSig-R, as well as a small user study showing that the proposed technique outperforms the expected human accuracy on the signature-verification task. Finally, a discussion on the impact of the number of signature examples (per writer) used during training on performance and execution time is presented.application/pdfporVerificação de assinaturaSoftwareSignature verificationOffline signature verificationWriter independent modelsAn offline writer independent signature verification method with robustness against scalings and rotationsinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisUniversidade Federal do Rio Grande do SulInstituto de InformáticaPrograma de Pós-Graduação em ComputaçãoPorto Alegre, BR-RS2022mestradoinfo:eu-repo/semantics/openAccessreponame:Biblioteca Digital de Teses e Dissertações da UFRGSinstname:Universidade Federal do Rio Grande do Sul (UFRGS)instacron:UFRGSTEXT001146961.pdf.txt001146961.pdf.txtExtracted Texttext/plain156490http://www.lume.ufrgs.br/bitstream/10183/247543/2/001146961.pdf.txta3823d652f5859187081885731f1046fMD52ORIGINAL001146961.pdfTexto completo (inglês)application/pdf1617064http://www.lume.ufrgs.br/bitstream/10183/247543/1/001146961.pdf466a85ae2bfceed8b184300e8f7327f6MD5110183/2475432022-08-21 04:40:15.116495oai:www.lume.ufrgs.br:10183/247543Biblioteca Digital de Teses e Dissertaçõeshttps://lume.ufrgs.br/handle/10183/2PUBhttps://lume.ufrgs.br/oai/requestlume@ufrgs.br||lume@ufrgs.bropendoar:18532022-08-21T07:40:15Biblioteca Digital de Teses e Dissertações da UFRGS - Universidade Federal do Rio Grande do Sul (UFRGS)false |
dc.title.pt_BR.fl_str_mv |
An offline writer independent signature verification method with robustness against scalings and rotations |
title |
An offline writer independent signature verification method with robustness against scalings and rotations |
spellingShingle |
An offline writer independent signature verification method with robustness against scalings and rotations Pachas, Felix Eduardo Huaroto Verificação de assinatura Software Signature verification Offline signature verification Writer independent models |
title_short |
An offline writer independent signature verification method with robustness against scalings and rotations |
title_full |
An offline writer independent signature verification method with robustness against scalings and rotations |
title_fullStr |
An offline writer independent signature verification method with robustness against scalings and rotations |
title_full_unstemmed |
An offline writer independent signature verification method with robustness against scalings and rotations |
title_sort |
An offline writer independent signature verification method with robustness against scalings and rotations |
author |
Pachas, Felix Eduardo Huaroto |
author_facet |
Pachas, Felix Eduardo Huaroto |
author_role |
author |
dc.contributor.author.fl_str_mv |
Pachas, Felix Eduardo Huaroto |
dc.contributor.advisor1.fl_str_mv |
Gastal, Eduardo Simões Lopes |
contributor_str_mv |
Gastal, Eduardo Simões Lopes |
dc.subject.por.fl_str_mv |
Verificação de assinatura Software |
topic |
Verificação de assinatura Software Signature verification Offline signature verification Writer independent models |
dc.subject.eng.fl_str_mv |
Signature verification Offline signature verification Writer independent models |
description |
Handwritten signatures are still one of the most used and accepted methods for user au thentication. They are used in a wide range of human daily tasks, including applications from banking to legal processes. The signature verification problem consists of verifying whether a given handwritten signature was generated by a particular person, by com paring it (directly or indirectly) to genuine signatures from that person. In this research work, a new offline writer-independent signature verification method is introduced (named VerSig-R), based on a combination of handcrafted Moving Least-Squares features and features transferred from a convolutional neural network. In our experiments, VerSig-R outperforms state-of-the-art techniques on Western-style signatures (CEDAR dataset), while also obtaining competitive results on South Asian-style handwriting (Bangla and Hindi datasets). Furthermore, a wide range of experiments demonstrate that VerSig-R is the most robust in relation to differences in scale and rotation of the signature images. This work also presents a discussion on dataset bias and on cross-dataset performance of VerSig-R, as well as a small user study showing that the proposed technique outperforms the expected human accuracy on the signature-verification task. Finally, a discussion on the impact of the number of signature examples (per writer) used during training on performance and execution time is presented. |
publishDate |
2022 |
dc.date.accessioned.fl_str_mv |
2022-08-20T04:55:47Z |
dc.date.issued.fl_str_mv |
2022 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10183/247543 |
dc.identifier.nrb.pt_BR.fl_str_mv |
001146961 |
url |
http://hdl.handle.net/10183/247543 |
identifier_str_mv |
001146961 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Biblioteca Digital de Teses e Dissertações da UFRGS instname:Universidade Federal do Rio Grande do Sul (UFRGS) instacron:UFRGS |
instname_str |
Universidade Federal do Rio Grande do Sul (UFRGS) |
instacron_str |
UFRGS |
institution |
UFRGS |
reponame_str |
Biblioteca Digital de Teses e Dissertações da UFRGS |
collection |
Biblioteca Digital de Teses e Dissertações da UFRGS |
bitstream.url.fl_str_mv |
http://www.lume.ufrgs.br/bitstream/10183/247543/2/001146961.pdf.txt http://www.lume.ufrgs.br/bitstream/10183/247543/1/001146961.pdf |
bitstream.checksum.fl_str_mv |
a3823d652f5859187081885731f1046f 466a85ae2bfceed8b184300e8f7327f6 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 |
repository.name.fl_str_mv |
Biblioteca Digital de Teses e Dissertações da UFRGS - Universidade Federal do Rio Grande do Sul (UFRGS) |
repository.mail.fl_str_mv |
lume@ufrgs.br||lume@ufrgs.br |
_version_ |
1800309201021435904 |