This study provides a writer-independent signature verification system. To classify the data, the system uses K-Nearest Neighbours (KNN) while Fourier Descriptors (FD) are used for feature extraction. In this case, to obtain reliable and steady features we were gathering, scanning and preparing signatures of ten people. As for the performance of the system, it achieved a 95% recognition rate on both the local and MCYT datasets where K=1. There is a need to develop something in this regards because it was shown that the misclassifications were due to having different signature limits. The findings reveal that both FD and KNN function well in writer independent model and provide a reliable solution to the problem of automated signature verification.
Article Details
Unique Paper ID: 166690
Publication Volume & Issue: Volume 11, Issue 2
Page(s): 1627 - 1637
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024