Nasaq-GP/Fonts

Arabic OCR Dataset

This dataset is designed for Arabic OCR tasks, aggregating data from multiple sources including the HICMA dataset.

Dataset Citation & Acknowledgements

A significant portion of this dataset is derived from the HICMA (Handwriting Identification of Manuscripts and Calligraphy in Arabic) dataset. We strictly adhere to their usage policy.

• Original Dataset: HICMA: Handwriting Identification of Manuscripts and Calligraphy in Arabic
• License: CC BY-NC 4.0 (Non-Commercial Use Only)

Citation

If you use this dataset, please cite the original HICMA paper:

@inproceedings{ismail-etal-2023-hicma,
    title = "{HICMA}: The Handwriting Identification for Calligraphy and Manuscripts in {A}rabic Dataset",
    author = "Ismail, Anis  and
      Kamel, Zena  and
      Mahmoud, Reem  and
      El Hajj, Hady",
    booktitle = "Proceedings of ArabicNLP 2023",
    month = dec,
    year = "2023",
    address = "Singapore (Hybrid)",
    publisher = "Association for Computational Linguistics",
    url = "[https://aclanthology.org/2023.arabicnlp-1.3](https://aclanthology.org/2023.arabicnlp-1.3)",
    pages = "24--32",
}