International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
Volume 57 - Issue 7 |
Published: November 2012 |
Authors: Tirtharaj Dash, Tanistha Nayak, Subhagata Chattopadhyay |
![]() |
Tirtharaj Dash, Tanistha Nayak, Subhagata Chattopadhyay . Handwritten Signature Verification (Offline) using Neural Network Approaches: A Comparative Study. International Journal of Computer Applications. 57, 7 (November 2012), 33-41. DOI=10.5120/9128-3295
@article{ 10.5120/9128-3295, author = { Tirtharaj Dash,Tanistha Nayak,Subhagata Chattopadhyay }, title = { Handwritten Signature Verification (Offline) using Neural Network Approaches: A Comparative Study }, journal = { International Journal of Computer Applications }, year = { 2012 }, volume = { 57 }, number = { 7 }, pages = { 33-41 }, doi = { 10.5120/9128-3295 }, publisher = { Foundation of Computer Science (FCS), NY, USA } }
%0 Journal Article %D 2012 %A Tirtharaj Dash %A Tanistha Nayak %A Subhagata Chattopadhyay %T Handwritten Signature Verification (Offline) using Neural Network Approaches: A Comparative Study%T %J International Journal of Computer Applications %V 57 %N 7 %P 33-41 %R 10.5120/9128-3295 %I Foundation of Computer Science (FCS), NY, USA
Forgery detection has been a challenging area in the field of biometry, e. g. , handwritten signatures. Signature verification is a bi-objective optimization problem. The two crucial parameters are accuracy and time of computation. In this work, a comprehensive study on application of Adaptive Resonance Theory (ART) Nets (Type 1 and 2) and Associative Memory Net (AMN) has been conducted. To decrease the time complexity a corresponding parallel version using OpenMP is developed for each algorithm. The algorithms are trained with the original/genuine signature and tested with a sample of twelve very similar-looking forged signatures. The study concludes that ART-1 detects fake signatures with an accuracy of 99. 89%; whereas, ART-2 and AMN detect forgery with accuracies of 99. 99% and 75. 68% respectively which are comparable to other methods cited in this paper.