31 Available online at www.ejournals.uofk.edu UofKEJ Vol. 8 Issue 1 pp.31-36 (February 2018) UNIVERSITY of KHARTOUM ENGINEERING JOURNAL (UofKEJ) Extended Feature Extraction Technique (Edge Direction Matrixes) For Online Arabic Handwriting Recognition Mohamed Mosadag Albadawi, Hozeifa Adam Abd Alshafy College of Computer Science and Information Technology, Karary University, Khartoum, Sudan. (Email:m.mosadag@gmail.com ) , (Email: hozeifa.adam@gmail.com ) _________________________________________________________________________________________________ ABSTRAC: Recognition of Arabic handwriting has attracted the interest of researchers for many years. Until now it has been a challenging research area due to many issues. The feature extraction is an essential stage in the recognition systems of handwriting. The main idea behind this paper is to study EDMs (Edge Direction Matrixes) as a feature extraction technique for Online Arabic Handwriting Recognition. In this study, SUSTOLAH datasets will be used, in which datasets of online Arabic handwriting are presented in Sudan University of Science and Technology. In this paper, satisfactory results have been achieved, where the value of the correlation/regress coefficient for the differences between the variant handwritten characters is found to be -0.01322. Keywords: On-line Arabic handwriting, feature extraction, edge direction matrixes (EDMS). _________________________________________________________________________________________________ INTRODUCTION In recent years, online handwriting recognition has become a central issue for Tablet PCs, PDAs screens, and hand-held computers. Online handwriting recognition is a task of mapping the handwriting words and characters to their text-based representations. In contrast to other languages, Arabic handwriting presents a lot of challenges [1]. Handwriting of the same characters can vary in both their sizes and styles even when written by the same person. The context- based difference in one person’s handwriting style presents another problem that faces Arabic handwriting recognition. Moreover, handwriting style has been affected by many factors such as the writer mood and the writing state. Great research effort has been put in the area of Arabic handwriting recognition [4] [6] [7] [11]. Nevertheless, limited progress has been achieved in this area. The Arabic language has 28 alphabet characters. Some of these characters take different formats depending on the character location within a word (beginning, middle or end) and they may differ in size. Moreover, there are many Arabic characters which can be associated with a single, double, or triple dots, or zigzag, As shown in Figure 1, the stages of pre-processing, feature extraction, and classification enact the key phases for the handwriting recognition systems. The Feature extraction phase has an important role in the task of the recognition. It acts to take the input pattern and transforms it to some data. This data is then used in the later recognition phases in order to categorize known or unknown patterns. The purpose of this paper is to study the EDMs as a feature extraction technique within the context of online Arabic handwriting recognition. We intend in this study to enhance the EDMs technique and then evaluate the enhanced technique with the use of some online Arabic handwriting characters which are taken from SUSTOLAH dataset. SUSTOLAH is a dataset of online Arabic handwriting which is presented in Sudan University of Science and Technology Online Arabic [10].