Offline Signature Recognition and Verification System Using Artificial Neural Network

  • Aqeel-ur-Rehman
  • Sadiq Ur Rehman
  • Zahid Hussain Babar
  • M. Kashif Qadeer
  • Faraz Ali Seelro
Keywords: Off-line Signature Verification, Image Preprocessing, Authentication

Abstract

There are several alternative life science techniques that are used to identify human, these techniques are namely eye recognition, face recognition, finger print recognition and currently a well-known signatures recognition and verification. The utilization of signatures is in all legal and financial documents. Verification of signatures now becomes necessary to distinguish between original and forged signature. A computer based technique is necessitated in this regard. Verification of signatures can be performed either offline or online. Under offline systems, signatures are taken as an image and recognition is performed using some artificial intelligence techniques including neural networks. We have worked on off-line Signature Recognition and Verification System (SRVS) by taking artificial neural network technique into account. Signatures are taken as image and after some necessary pre-processing (i.e. to isolate the signature area) training of system is done with some initial stored samples to which authentication is needed. MATLAB has been used to design the system. The system is tested for several scanned signatures and the results are found satisfactory (about approximately 95% success rate). Image quality plays an important role as poor quality of signature image may lead to the failure to recognize/verify a signature. Increase in the attributes/ features of signature will increase the verification ability of the system but it may lead to higher computational complexity

Published
2018-01-31
How to Cite
Aqeel-ur-Rehman, Sadiq Ur Rehman, Babar, Z. H., Qadeer, M. K., & Seelro, F. A. (2018). Offline Signature Recognition and Verification System Using Artificial Neural Network. University of Sindh Journal of Information and Communication Technology , 2(1), 73-80. Retrieved from https://sujo.usindh.edu.pk/index.php/USJICT/article/view/512

Most read articles by the same author(s)

Obs.: This plugin requires at least one statistics/report plugin to be enabled. If your statistics plugins provide more than one metric then please also select a main metric on the admin's site settings page and/or on the journal manager's settings pages.