The automatic recognition of text on scanned images has enabled many applications such as searching for words in large volumes of documents, automatic sorting of postal mail, and convenient editing of previously printed documents. The domain of handwriting in the Arabic script presents unique technical challenges and has been addressed more recently than other domains. Many different methods have been proposed and applied to various types of images. Here we will focus on the recognition part of handwritten Arabic letters and digits recognition that face several challenges, including the unlimited variation in human handwriting and the large public databases.
In this project we built a model which can classify a new image to an arabic letter or digit and get accuracy of 98.86% when testing on more than 13000 different images.
You can download the dataset from here
- You need to install Jupyter Notebook or use Colab to run the iPython Noteboob direclty.
- You should also have python worked at version 2.6 or higher.
- You should install Keras (follow the documentation here).
This project is licensed under the MIT License - see the LICENSE.md file for details