Hala Bezine IGS 2011 - Hala Bezine IGS 2011 - LIVE AQUA CANCUN, MEXICOLIVE AQUA CANCUN, MEXICO 1

Research Group on Intelligent Machines

Prediction of Psychophysical Laws with the Beta-Elliptic Model

Hala Bezine Adel M. Alimi

hala.bezine@ieee.org adel.alimi@ieee.org

National School of Engineers of Sfax

REsearch Group on Intelligent Machines

Extraction of the curvilinear velocity signal V(t)

Generation of the Beta profiles with the kinematic parameters (t0, tc, t1, p, q, H) for each stroke

Extraction of the associated static parameters (a, b, X0, Y0, θ0) related to the elliptic shape

Regeneration of the original handwriting script with Beta-elliptic strokes

On-line handwriting script

Production of handwriting requires a hierarchically organized flow of

information undergoing a series of transformations.

Extraction of a simple stroke (a) in the velocity domain (b) in the trajectory domain

Hala Bezine IGS 2011 - Hala Bezine IGS 2011 - LIVE AQUA CANCUN, MEXICOLIVE AQUA CANCUN, MEXICO

Example of Arabic letter « س » generated for three times with variable velocity and constant shape, with the respective

curvilinear velocity profiles Vx and Vy

The amount of time of the writing task decreases, if the speed increases.

Example of Arabic letter «و »generated for three times with constant velocity and variable shape

The model yields to the isochrony principle, and the writing time is not unchanged across changes in writing size, but increases by a small amount

The relation between the dynamics and the shape of a

handwriting movement Use spiking neural networks.

An elliptic trajectory leads to a velocity signal that can be fitted

with a Beta profile.

For different conditions most psychophysical laws are verified

such as isochrony, speed and size scaling.

Prediction of psychophysical laws with the Beta-elliptic model