Integration of Bayesian Classifier and Perceptron for Problem Identification on Dynamics Signature Using a Genetic Algorithm for the Identification Threshold Selection

Kostyuchenko, Evgeny; Gurakov, Mihail; Krivonosov, Egor; Tomyshev, Maxim; Mescheryakov, Roman; Hodashinskiy, Ilya

doi:10.1007/978-3-319-40663-3_71

Cited by 5 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It was used to form the following dynamic sequences of the discrete time: 1) pen location relatively to the x, y, z axes; 2) pen pressure P; 3) azimuth α; 4) pen height angle θ relatively to the tablet. The observation dataset was created from the first eight harmonics of the specified sequences' decomposition into a Fourier series [34]. Thus, each entry of the observation table was a description of the handwritten signature with 144 features and the class mark being the signing user's number.…”

Section: Methodsmentioning

confidence: 99%

Gravitational Search For Designing A Fuzzy Rule-Based Classifiers For Handwritten Signature Verification

Bardamova

Konev

Hodashinsky

et al. 2019

JCOMSS

View full text Add to dashboard Cite

Handwritten signatures are used in authentication systems as a universal biometric identifier. Signature authenticity verification requires building and training a classifier. This paper describes a new approach to the verification of handwritten signatures by dynamic characteristics with a fuzzy rule-based classifier. It is suggested to use the metaheuristic Gravitational Search Algorithm for the selection of the relevant features and tuning fuzzy rule parameters. The efficiency of the approach was tested with an original dataset; the type II errors in finding the signature authenticity did not exceed 0.5% for the worst model and 0.08% for the best model.

show abstract

Section: Methodsmentioning

confidence: 99%