Online signature recognition using Kekre's vector quantization algorithms KMCG & KFCG

  • Authors:
  • H. B. Kekre;V. A. Bharadi;V. I. Singh;S. Gupta;N. Kulkarni

  • Affiliations:
  • MPSTME, NMIMS University, Mumbai, India;MPSTME, NMIMS University, Mumbai, India;TCET, Mumbai University, Mumbai, India;TCET, Mumbai University, Mumbai, India;TCET, Mumbai University, Mumbai, India

  • Venue:
  • Proceedings of the International Conference & Workshop on Emerging Trends in Technology
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Dynamic Signature Recognition is one of the highly accurate biometric traits. We capture live signature of the person hence it is possible to analyze dynamic characteristics of signature for matching purpose. The signature captured by digitizer gives information about dynamic nature of signature and pressure applied while signing. Dynamic parameters such as pressure, X, Y, Z- co-ordinates, Azimuth & Altitude are captured. These signature points are vector in n-dimensional vector space. In this paper we have proposed clustering of these points in vector space to form feature vector is proposed for online signature recognition. For clustering & codebook generation kekre's Vector Quantization Algorithms such as KFCG, KMCG are used with variations. The proposed technique gives up to 97% accuracy.