Extraction of delta-lognormal parameters from handwriting strokes

Abstract

In the context of the Kinematic Theory of Rapid Human Movement, handwriting strokes are considered to be primitives that reflect the intrinsic properties of the neuromuscular system of a writer as well as the basic control strategies that the writer uses to produce such strokes. The study of these strokes relies on the extraction of the different parameters that characterize a stroke velocity profile. In this paper, we present a new method for stroke parameter extraction. The algorithm is described and evaluated under various testing conditions.

Keywords

Kinematic Theory, lognormal curve, neuromuscular systems, handwriting, strokes, velocity profiles, parameter estimation, non-linear regression, optimization