Anova-based feature analysis and selection in HMM-based offline signature verification system

• Main Authors: Balbed M.A.M., Ahmad S.M.S., Shakil A.
• Other Authors: 24721384800
• Format: Conference Paper
• Published: 2023
• Subjects: Hidden Markov models; Industrial applications; Intelligent systems; Pixels; Regression analysis; Analysis techniques; Center of gravity; Centre of gravity; Density features; Feature analysis; Off-line signature verification; Offline; Skilled forgery; Analysis of variance (ANOVA)

This paper presents an analysis performance of different features in distinguishing between genuine and forged signatures for HMM based offline signature verification systems. The four offline features include pixel density, centre of gravity, distance and angle. All features considered are local in nature. The analysis technique used here is based on analysis of variance (ANOVA). Experimental results show that the combination of center of gravity and pixel density features are good for distinguishing between genuine and skilled forgeries for an HMM based offline signature verification system. � 2009 IEEE.