Home Publications Teaching Resources Contacts

The AIA9k Dataset for Arabic Handwritten Character Recognition

This dataset is part of a research project, pending availability of further funding. The current version of the dataset contains 8737 samples of the 28 Arabic alphabet letters. The samples were collected from 107 writers.


You can download the dataset from here.

How to Cite This Annotation?

We hope you find this dataset useful. In case you use it, please cite the following paper.

Marwan Torki, Mohamed E. Hussein, Ahmed Elsallamy, Mahmoud Fayyaz, and Shehab Yaser, "Window-Based Descriptors for Arabic Handwritten Alphabet Recognition: A Comparative Study on a Novel Dataset," arXiv:1411.3519 [cs.CV], November 2014. [pdf][bibtex][data]