Cover Image

A novel architecture to verify offline hand-written signatures using convolutional neural network

Hand-written signatures are marked on documents to establish legally binding evidence of identity and intent. However, they are prone to forgery, and the design of an accurate feature extractor to distinguish between highly-skilled forgeries and genuine signatures is a challenging task. In this pape...

Full description

Saved in:
Bibliographic Details
Main Authors: Alkaabi, S., Yussof, S., Almulla, S., Al-Khateeb, H., Alabdulsalam, A.A.
Format: Conference Paper
Language:English
Published: 2020
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uniten.dspace-12874
record_format dspace
spelling my.uniten.dspace-128742020-07-07T06:21:50Z A novel architecture to verify offline hand-written signatures using convolutional neural network Alkaabi, S. Yussof, S. Almulla, S. Al-Khateeb, H. Alabdulsalam, A.A. Hand-written signatures are marked on documents to establish legally binding evidence of identity and intent. However, they are prone to forgery, and the design of an accurate feature extractor to distinguish between highly-skilled forgeries and genuine signatures is a challenging task. In this paper, we propose a Convolution Neural Network (CNN) architecture for Signature Verification (SV). The algorithm is trained using two signatures, genuine and forged. Then the SV module performs a classification task to determine if any two signatures are of the same individual or not. The simulation results show that the proposed method can achieve 27% (relatively) better results than the benchmark scheme. The paper also integrated different data augmentation techniques for the signature data, which further improved the efficiency of the proposed method by 14% (relative). © 2019 IEEE. 2020-02-03T03:27:28Z 2020-02-03T03:27:28Z 2019 Conference Paper 10.1109/3ICT.2019.8910275 en
institution Universiti Tenaga Nasional
building UNITEN Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Tenaga Nasional
content_source UNITEN Institutional Repository
url_provider http://dspace.uniten.edu.my/
language English
description Hand-written signatures are marked on documents to establish legally binding evidence of identity and intent. However, they are prone to forgery, and the design of an accurate feature extractor to distinguish between highly-skilled forgeries and genuine signatures is a challenging task. In this paper, we propose a Convolution Neural Network (CNN) architecture for Signature Verification (SV). The algorithm is trained using two signatures, genuine and forged. Then the SV module performs a classification task to determine if any two signatures are of the same individual or not. The simulation results show that the proposed method can achieve 27% (relatively) better results than the benchmark scheme. The paper also integrated different data augmentation techniques for the signature data, which further improved the efficiency of the proposed method by 14% (relative). © 2019 IEEE.
format Conference Paper
author Alkaabi, S.
Yussof, S.
Almulla, S.
Al-Khateeb, H.
Alabdulsalam, A.A.
spellingShingle Alkaabi, S.
Yussof, S.
Almulla, S.
Al-Khateeb, H.
Alabdulsalam, A.A.
A novel architecture to verify offline hand-written signatures using convolutional neural network
author_facet Alkaabi, S.
Yussof, S.
Almulla, S.
Al-Khateeb, H.
Alabdulsalam, A.A.
author_sort Alkaabi, S.
title A novel architecture to verify offline hand-written signatures using convolutional neural network
title_short A novel architecture to verify offline hand-written signatures using convolutional neural network
title_full A novel architecture to verify offline hand-written signatures using convolutional neural network
title_fullStr A novel architecture to verify offline hand-written signatures using convolutional neural network
title_full_unstemmed A novel architecture to verify offline hand-written signatures using convolutional neural network
title_sort novel architecture to verify offline hand-written signatures using convolutional neural network
publishDate 2020
_version_ 1672614185680240640
score 13.222552

Similar Items