Abstract—Signature verification is an actively researched area whose goal is to decide whether unknown signatures are genuine or forged. Online signature verification applies signatures captured with an electronic device (digital tablet or pen). Online signatures contain not only spatial information but dynamics as well. There are two types of possible errors, the false prediction as genuine and the false prediction as a forgery. This paper proposes a prediction score as the classification output, which indicates the confidence of the system decision. This approach allows a trade-off between the different error types to create specialized verifiers and construct combined classifiers. This paper presents two types of combined classifiers, pre-filtering classifiers and majority voting classifiers. The proposed approaches are evaluated using the MCYT-100 dataset.
 

Index Terms—online signature verification, dynamic time warping, DTW, confidence score, prediction score, nonbinary decision, ensemble classification

Cite: András Heszler, Cintia Lia Szücs, and Bence Kővári, "The Use of Confidence Indicating Prediction Score in Online Signature Verification," Journal of Advances in Information Technology, Vol. 13, No. 3, pp. 290-294, June 2022.

Copyright © 2022 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.