ONLINE HANDWRITING SIGNATURE: AN EXPERIMENTAL STUDY ON PRESSURE SIGNAL NORMALIZATION IN DIFFERENT DEVICES FOR BIOMETRIC ACQUISITION
Price
Free (open access)
Transaction
Volume
206
Pages
10
Page Range
175 - 184
Published
2022
Paper DOI
10.2495/SAFE210151
Copyright
Author(s)
FABIO GARZIA, MARIA CHIARA GIANNONE, MARCOS FAUNDEZ-ZANUY
Abstract
Online handwriting signature is an important element in biometrics and therefore for security applications. In this paper the characterization of pressure for different acquisition devices for digital signature is presented, considering that technological evolution always produces new devices and that is why it is necessary to facilitate interoperability between them. The devices used are three different Wacom tablets: Wacom Intuos pro, Wacom Cintiq 16 and STU-530. Each of them has different pens and more than one nib and that is why for each of them over one case was analyzed in reference to the different pens used. It was experimented how, thanks to the respective software of the tablets, it is possible to acquire the pressure that is generated by the pen on the tablet and to connect it to the weight force exerted by them thanks to a balance placed under the tablet. The process used allowed to create for each tablet the pressure/weight plot that permitted to derive the different sensitivities of the tablets to pressure and also the different saturation values. In addition, it was also analyzed the variation of the thickness of the stroke exerted by the pen going to write on the tablet with different values of force applied. Thanks to the use of carbon copy paper, which allowed to reproduce the line generated by the pen on a sheet of paper, it was then possible to study under the microscope the trend of the thickness of the line drawn as the force varies. The obtained results allowed to obtain a direct comparison between the signals acquired by different devices thanks to the normalization processes that permit to increase the recognition performance.
Keywords
online handwriting signature, pressure, software, thickness, normalization processes