An open API service indexing awesome lists of open source software.

https://github.com/bertsky/mkn-kurrent-gt

Kurrent GT from the Moravian Knowledge Network handwritten periodicals
https://github.com/bertsky/mkn-kurrent-gt

Last synced: 5 months ago
JSON representation

Kurrent GT from the Moravian Knowledge Network handwritten periodicals

Awesome Lists containing this project

README

          


mkn-kurrent-gt


Kurrent GT from the Moravian Knowledge Network handwritten periodicals


Metadata



Language:

deu

Format:

Page-XML

Time:

1765-1806

GT Type:

data_structure_and_text

License:

PublicDomainMark 1.0

Transcription Guidelines:

OCR-D GT Level 2, but with ℓ as

Project:

Gemein-Nachrichten der Herrnhuter Brüdergemeine

Project-URL:

https://dhh.hypotheses.org/


Sources


The volume of transcriptions:





TextLine
Page
TxtRegion




19606
771
1220




List of transcriptions






document
TxtRegion
ImgRegion
LineDrawRegion
GraphRegion
TabRegion
ChartRegion
SepRegion
MathRegion
ChemRegion
MusicRegion
AdRegion
NoiseRegion
UnkownRegion
CustomRegion
TextLine
Page




GN_1770_2_GN_A_148
239













2984
107


GN_1773_3_GN_A_167
66













1131
42


GN_1774_2_GN_A_170
19













387
13


GN_1765_1_GN_A_109
4













32
1


GN_1771_4_GN_A_158
42













641
23


GN_1765_2_GN_A_110
51













728
23


GN_1806_4_GN_A_354
136













2422
102


GN_1807_1_GN_A_355
3













24
1


GN_1806_1_GN_A_351
445













8723
358


GN_1774_3_GN_A_171
26













570
20


GN_1788_5_GN_A_250
189













1964
81






Extent



This Ground Truth contains a random subsample of pages from various
issues of the _Gemein-Nachrichten_ of the Moravian congregation
(Brüder-Gemeine) from Herrnhut. The German texts are written in Kurrent
by various writers and originate between 1765 and 1807.
The METS files derive directly from the respective digitised objects in the
Digital Collections of SLUB.
They have been completed by an additional fileGrp `GT-PAGE`.



For workflows how these files were created and can be used for training
OCR models, cf. Wiki pages.