Arabic Handwriting Word Recognition Based on Scale Invariant Feature Transform and Support Vector Machine

  • Alia Karim Abdul Hassan Computer Science Department, University of Technology, Baghdad, Iraq
  • Bashar Saadoon Mahdi Computer Science Department, University of Technology, Baghdad, Iraq
  • Asmaa Abdullah Mohammed Computer Science Department, University of Technology, Baghdad, Iraq
Keywords: AHDB Database, SIFT feature extraction, SVM classifier algorithm

Abstract

Offline Arabic handwritten recognition lies in a major field of challenge due to the changing styles of writing from one individual to another. It is difficult to recognize the Arabic handwritten because of the same appearance of the different characters.  In this paper a proposed method for Offline Arabic handwritten recognition. The   proposed method for recognition hand-written Arabic word without segmentation to sub letters based on feature extraction scale invariant feature transform (SIFT) and   support vector machines (SVMs) to enhance the recognition accuracy. The proposed method  experimented using (AHDB) database. The experiment result  show  (99.08) recognition  rate.

Published
2019-02-28
How to Cite
Abdul Hassan, A. K., Mahdi, B. S., & Mohammed, A. A. (2019). Arabic Handwriting Word Recognition Based on Scale Invariant Feature Transform and Support Vector Machine. Iraqi Journal of Science, 60(2), 381-387. Retrieved from https://ijs.uobaghdad.edu.iq/index.php/eijs/article/view/543
Section
Computer Science

Most read articles by the same author(s)