Some remarks on a set of information theory features used for on-line signature verification

Recently a new set consisting in six information theory features was proposed to on-line signature verification by Rosso, Ospina and Frery . The proposed features were evaluated on the MCYT-100 on-line signature database resulting in the best performance ever measured on that dataset. In this paper we repeat their measurements and show that their result is erroneous. In addition, we evaluate the performance of the same on-line signature verification system using exactly the same number of state-of-the-art features. State-of-the-art features always outperform the information theory related features, regardless of the used classification method.



State-of-the-art feature set (rf369)

Information theory feature set

R script for SVM evaluation (László Zsolt SZABÓ)

Results

Verification performance results for the two feature sets and skilled forgery case are depicted in the form of ROC plots in the following figures.

Publication

Margit ANTAL and László Zsolt SZABÓ, Some remarks on a set of information theory features used for on-line signature verification, 2017 5th International Symposium on Digital Forensic and Security (ISDFS), Tirgu Mures, 2017, pp. 1-5. doi: 10.1109/ISDFS.2017.7916498 [LINK] [Download PDF].