Bio-inspired reference level assigned DTW for person identification using handwritten signatures

Authors:
Muzaffar Bashir;Jürgen Kempf
Affiliations:
Faculty of Electronics and Information Technology, University of Applied Sciences Regensburg, Germany;Faculty of Electronics and Information Technology, University of Applied Sciences Regensburg, Germany
Venue:
BioID_MultiComm'09 Proceedings of the 2009 joint COST 2101 and 2102 international conference on Biometric ID management and multimodal communication
Year:
2009

Citing 3
Cited 1

Scaling up dynamic time warping for datamining applications

Proceedings of the sixth ACM SIGKDD international conference on Knowledge discovery and data mining
Online signature verification using a new extreme points warping technique

Pattern Recognition Letters
Exact indexing of dynamic time warping

Knowledge and Information Systems

DTW Based Classification of Diverse Pre-Processed Time Series Obtained from Handwritten PIN Words and Signatures

Journal of Signal Processing Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Person identification or verification is becoming an important issue in our life with the growth of information technology. Handwriting features of individuals during signing are used as behavioral biometrics. This paper presents a new method for recognizing person using online signatures based on reference level assigned Dynamic Time Warping (DTW) algorithm. The acquisition of online data is carried out by a digital pen equipped with pressures and inclination sensors. The time series obtained from pen during handwriting provide valuable insight to the unique characteristics of the writers. The obtained standard deviation values of time series are found person specific and are used as so called reference levels. In the proposed method reference levels are added to time series of query and sample dataset before dynamic time warping distance calculations. Experimental results show that the performance of accuracy in person authentication is improved and computational time is reduced.