Overview

The MS-TEx 2015 dataset contains 10 handwritten and machine-printed historical document images along with eight spectral images for each image.
The goal of this contest is evaluation of the most recent advances in text extraction from historical document images captured by a multispectral imaging system.

Citation

@INPROCEEDINGS{6628610,
    author={Hedjam, R. and Cheriet, M.},
    booktitle={Document Analysis and Recognition (ICDAR), 2013 12th
    International Conference on},
    title={Ground-Truth Estimation in Multispectral Representation Space:
    Application to Degraded Document Image Binarization},
    year={2013},
    month={Aug},
    pages={190-194},
    doi={10.1109/ICDAR.2013.45},
    ISSN={1520-5363}
}

@article{DBLP:journals/pr/HedjamC13,
    author = {Rachid Hedjam and
    Mohamed Cheriet},
    title = {Historical document image restoration using multispectral imaging
    system},
    journal = {Pattern Recognition},
    volume = {46},
    number = {8},
    pages = {2297--2312},
    year = {2013},
    url = {http://dx.doi.org/10.1016/j.patcog.2012.12.015},
    doi = {10.1016/j.patcog.2012.12.015},
    timestamp = {Mon, 08 Apr 2013 20:24:35 +0200},
    biburl = {http://dblp.uni-trier.de/rec/bib/journals/pr/HedjamC13},
    bibsource = {dblp computer science bibliography, http://dblp.org}
}