On the use of first and second derivative approximations for biometric online signature recognition

TL;DR

This study evaluates how different derivative approximation methods affect feature extraction in online signature recognition, demonstrating that 11-point approximation improves identification accuracy and reduces forgery success rates.

Contribution

It introduces a comparative analysis of approximation methods for delta features, highlighting the superior performance of 11-point approximation in biometric signature recognition.

Findings

11-point approximation outperforms 1-point approximation

1.4% increase in identification rate

36.8% reduction in random forgeries

Abstract

This paper investigates the impact of different approximation methods in feature extraction for pattern recognition applications, specifically focused on delta and delta-delta parameters. Using MCYT330 online signature data-base, our experiments show that 11-point approximation outperforms 1-point approximation, resulting in a 1.4% improvement in identification rate, 36.8% reduction in random forgeries and 2.4% reduction in skilled forgeries