Sindhi OCR Dataset: Multi-Font Machine-Printed Text Line Images with Ground Truth for Low-Resource Language Recognition
收藏官方服务：
资源简介：
This dataset has 60,122 binary TIF text line images across different Sindhi fonts. Ground truth provided as UTF-8 encoded .txt files and .xml metadata files. Train/test splits provided as split files for reproducibility. For academic and non-commercial research use only.
提供机构：
Zenodo创建时间：
2026-05-18
