Computer Vision Lab, TU Wien
Published June 27, 2019 | Version v2
Dataset Open

MSBin - MultiSpectral Document Binarization

Creators

  • 1. Computer Vision Lab, TU Wien

Description

This dataset is named MSBin which stands for MultiSpectral Document Binarization. The dataset is dedicated to the (document image) binarization of multispectral images. A ReadMe is contained within the dataset and also available here:
ReadMe

The dataset is introduced in:
Fabian Hollaus, Simon Brenner, Robert Sablatnig: CNN Based Binarization of MultiSpectral Document Images. ICDAR 2019: 533-538

Notes

Note that this is the second version of the dataset, where 10 images are removed from the test set, because they were too varying from the training set. The results obtained on the first version can be found in [Hollaus et al. 2019]. Recent results are given in the ReadMe.

Files

MSBin_v2.zip

Files (1.7 GB)

Name Size Download all
md5:ae9088f2a4e7c8e7bae7f7f0c777ff3b
1.7 GB Preview Download

Additional details

Funding

European Commission
READ - Recognition and Enrichment of Archival Documents 674943
FWF Austrian Science Fund
The Origin of the Glagolitic-Old Church Slavonic Manuscripts P 29892