Efficient Preprocessing Algorithm for Online Handwritten Arabic Strokes

Abstract

Developing an online handwriting recognition system for Arabic script used in pen-based devices plays an important role in making these devices available and usable for the Arabic society. This paper develops a preprocessing algorithm for strokes written in Arabic script to improve the strokes recognition process, since there are some troubles on distinguishing the written stroke from similar characters. The uniqueness of the recommended preprocessing algorithm is the dealing with every stroke in the character separately. The algorithm is comprised of five essential phases that will settle on the strokes prepared and make the strokes ready for the next phase in recognition. Before the preprocessing algorithm, a basic step called stroke capturing is done, where data points are collected along the path of an input device (stylus pen or mouse) same time those characters are drawn.