Skip to content

Latest commit

 

History

History
8 lines (5 loc) · 470 Bytes

README.md

File metadata and controls

8 lines (5 loc) · 470 Bytes

ArabicOCR

This project is a group project done by 4 undergrads in Cairo University, the project is a collection of jupyter notebooks that take a scanned image of arabic text and then separate it into words then into characters then classifies them into arabic characters and applies post processing methods to make our accuracy better.

OCR_MainProgramFinal is the complete script, it takes the model and the scanned image directory and produces the files of text.