Open AccessKernel Based K-Nearest Neighbor Method to Enhance the Performance and Accuracy of Online Signature Recognition
Author(s) -
R. Ravi Chakravarthi,
E. Chandra
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.j9139.0881019
Subject(s) - k nearest neighbors algorithm , biometrics , computer science , pattern recognition (psychology) , signature (topology) , word error rate , artificial intelligence , kernel (algebra) , signature recognition , process (computing) , key (lock) , data mining , mathematics , computer security , geometry , combinatorics , operating system
Signature recognition is a significant among the most fundamental biometrics recognition techniques, is a key bit of current business works out, and is considered a noninvasive and non-undermining process. For online signature recognition, numerous methods had been displayed previously. In any case, accuracy of the recognition framework is further to be enhanced and furthermore equal error rate is further to be decreased. To take care of these issues, a novel order method must be proposed. In this paper, Kernel Based k-Nearest Neighbor (K-kNN) is presented for online signature recognition. For experimental analysis, two datasets are utilized that are ICDAR Deutsche and ACT college dataset. Simulation results show that, the performance of the proposed recognition technique than that of the existing techniques in terms of accuracy and equal error rate. Keywords: Online signature recognition, Kernel Based kNearest Neighbor (K-kNN), accuracy, equal error rate.