Arabic (Indian) Handwritten‏ ‏Digits Recognition Using Multi feature and KNN Classifier

Main Article Content

Alia Karim Abdul Hassan

Abstract

This paper presents an Arabic (Indian)  handwritten digit recognition system based on combining  multi feature  extraction methods, such a upper_lower  profile, Vertical _ Horizontal projection and Discrete Cosine Transform (DCT) with Standard Deviation σi called (DCT_SD)  methods. These  features are extracted from the image  after dividing it by several blocks. KNN classifier used  for classification purpose. This work is tested with the ADBase standard database (Arabic numerals),  which consist of 70,000 digits were 700 different writers write  it. In proposing system used 60000 digits, images for training phase and 10000 digits, images in testing phase. This work  achieved  97.32%  recognition  Accuracy

Abstract Views: 0 | PDF Downloads: 2

Article Details

How to Cite
Abdul Hassan, A. (2018). Arabic (Indian) Handwritten‏ ‏Digits Recognition Using Multi feature and KNN Classifier. Journal of University of Babylon, 26(4), 10-17. https://doi.org/https://doi.org/10.29196/jub.v26i4.679
Section
Articles