Can AI Finally Crack Ottoman Text Recognition?

This article reviews research on OCR for machine-typed historical documents, with a focus on Arabic, Persian, and Ottoman scripts. While no system can yet accurately extract text from old Kurdish publications in Arabic-Persian script, related studies show progress through neural networks, SVMs, and deep learning. From early models with limited accuracy to modern convolutional networks achieving near 98% recognition, the literature highlights both the challenges of cursive scripts and the potential of AI-driven approaches for digitizing heritage texts.

Table of Links

Abstract and 1. Introduction

1.1 Printing Press in Iraq and Iraqi Kurdistan

1.2 Challenges in Historical Documents

1.3 Kurdish Language

2 Related work

This section reviews the literature by focusing on machine-typed historical documents. To the best of our knowledge, currently, there is no OCR system that can accurately extract text from old Kurdish publications written in Arabic-Persian script. Therefore ...