On-line recognition of handwritten Arabic characters

Arabic characters are always in cursive script. Handwritten words were entered into an IBM PC via a graphics tablet and a segmentation process applied to the points; the length and the slope of each segment was then found, and the slope categorized into one of four directions. In the learning process, specifications on the strokes of each character are fed to the computer. In the recognition process, the parameters of each stroke are found and special rules applied to select the collection of strokes which best matches the features of one of the stored characters. The results are promising, and suggestions for improvements leading to 100% recognition are proposed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>