README.md

Text Extraction and Restoration of Old Handwritten Documents

Arxiv Paper

Introduction

This paper provides a CNN based approach to old handwritten document image binarization. We introduce a newly collected ISI-letters dataset which is collected from the archives of Indian Statistical Institute.

Requirements : python2

If you use this code consider citing:

@article{wadhwani2021text,
  title={Text extraction and restoration of old handwritten documents},
  author={Wadhwani, Mayank and Kundu, Debapriya and Chakraborty, Deepayan and Chanda, Bhabatosh},
  journal={Digital Techniques for Heritage Presentation and Preservation},
  pages={109--132},
  year={2021},
  publisher={Springer}
}