A fuzzy genetic system for segmentation of on-line handwriting: Application to ADAB database

Abstract

In this paper, we present a fuzzy genetic system for handwriting segmentation via perceptual codes based on assumptions of PerTOHS theory, which correspond to: handwriting is a form and a sequence of perceptual codes. Studying handwriting, we perceive the existence of elementary and global perceptual codes. Gathering the elementary ones in a choice of constraints we obtain global ones, and to obtain different forms of handwriting, we proceed by a perceptual organization of them. So, we present a new fuzzy genetic system to improve handwriting segmentation via perceptual codes. This system is based on the Beta-elliptic model, uses the fuzzy set theory to detect the elementary perceptual codes (EPCs) and the genetic algorithms for the global perceptual ones (GPCs). To validate our new system, we use ADAB database. The obtained results show successful representations of handwritten script via perceptual codes.