Data

Tools

Indexes

Applications

Experiments

Open Source Science

ocr-data

https://github.com/millawell/ocr-data

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (8.7%) to scientific vocabulary
Last synced: 10 months ago · JSON representation ·

Repository

Basic Info
  • Host: GitHub
  • Owner: millawell
  • License: apache-2.0
  • Language: Jupyter Notebook
  • Default Branch: master
  • Size: 2.49 MB
Statistics
  • Stars: 6
  • Watchers: 2
  • Forks: 4
  • Open Issues: 4
  • Releases: 2
Created almost 6 years ago · Last pushed almost 3 years ago
Metadata Files
Readme License Citation

README.md

ocr-data

This repository consists of

  • a ground truth OCR data set for historical prints from around 1830.
  • a framework to create and share your own ground truth OCR data sets if you don't own the copyright for the images used.

How to get the ground truth OCR data set?

The data set can be found in the data directory and consists of a METS file for each of the PDFs that were used for transcription and a directory data/page_xml that contains the transcriptions of the ground truth in PAGE-XML format. The data is published under a CC-BY license (data/LICENSE).

The PDFs are not hosted here, but have to be retrieved from the respective institutions and can then be combined with the transcriptions found here. To compile the data set, please

  • download all PDFs listed in the *.mets files into the data/pdf_renamed/ directory and rename them ${identifier}.pdf
  • change to the pipelines directory and run the make command

How to create your own ground truth OCR data set?

  • Collect a set of PDFs from Google Books or the Internet Archive and select a set of pages that you would like to transcribe
  • transcribe the text on the images for each PDF individually with the ketos transcribe framework found here http://kraken.re/ketos.html (Kiessling 2019) and store the resulting *.html in a directory named after the PDFs identifier within the data/transcriptions directory.
  • Now, you can run python create_xml_files.py for each of the PDFs which will output a data set similar to the one from our case study in this repository and other scholars who would like to use your data set can reproduce it without you having to publish the Google Books PDF yourself.

The source code is published under an Apache License (LICENSE).

Kiessling, Benjamin. “Kraken - an Universal Text Recognizer for the Humanities.” DH Conference Proceedings, vol. 30, 2019.

Owner

  • Login: millawell
  • Kind: user

Citation (citation.cff) 

cff-version: 1.2.0
message: "If you use the data set, please cite it as below."
authors:
- family-names: "Lassner"
  given-names: "David"
  orcid: "https://orcid.org/0000-0001-9013-0834"
- family-names: "Coburger"
  given-names: "Julius"
- family-names: "Neudecker"
  given-names: "Clemens"
  orcid: "https://orcid.org/0000-0001-5293-8322"
- family-names: "Baillot"
  given-names: "Anne"
  orcid: "https://orcid.org/0000-0002-4593-059X"
title: "Data set of the paper "Publishing an OCR ground truth data set for reuse in an unclear copyright setting""
version: 1.1.0
doi: 10.5281/zenodo.4742068
date-released: 2021-05-07
url: "https://github.com/millawell/ocr-data"

GitHub Events

Total
Last Year

Issues and Pull Requests

Last synced: over 1 year ago

All Time
  • Total issues: 4
  • Total pull requests: 2
  • Average time to close issues: N/A
  • Average time to close pull requests: 11 minutes
  • Total issue authors: 3
  • Total pull request authors: 2
  • Average comments per issue: 0.75
  • Average comments per pull request: 0.0
  • Merged pull requests: 1
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 2
  • Pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Issue authors: 1
  • Pull request authors: 0
  • Average comments per issue: 0.0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • alix-tz (1)
  • geeky-programer (1)
  • cneud (1)
Pull Request Authors
  • PonteIneptique (1)
  • cneud (1)
Top Labels
Issue Labels
Pull Request Labels