A novel comprehensive database for arabic and english off-line handwritten digits recognition

Ahmed Subhi Abdalkafor, Waleed Kareem Awad, Khattab M. Ali Alheeti

Abstract


The recognition of Arabic handwritten is received at the same interest as other Latin languages. In optical character recognition (OCR), handwriting Arabic recognition is considered as one of the critical and difficult tasks in the various scientific area. The main issues of this matter were due to the lack of public Arabic handwriting databases and the cursive nature of Arabic writing. In this paper, a new benchmark database is built for the Arabic and English off-line handwritten digits Recognition. The original form is divided into three groups: Arabic digits, English digits, and word Arabic digits which written five times by 100 different academic staff and students of university writers. Our database contains 14500 images; divided into two subsets of training and testing to help researchers through evaluating and comparing results obtained from their systems. 

Keywords


Database; Arabic; Handwritten; Word arabic digits; Cropping.

Full Text:

PDF


DOI: http://doi.org/10.11591/ijeecs.v20.i1.pp145-149

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Indonesian Journal of Electrical Engineering and Computer Science (IJEECS)
p-ISSN: 2502-4752, e-ISSN: 2502-4760
This journal is published by the Institute of Advanced Engineering and Science (IAES) in collaboration with Intelektual Pustaka Media Utama (IPMU).

shopify stats IJEECS visitor statistics