books/quixote-shelton-ocr
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Activity Actions
Scripts to do a high quality OCR of Thomas Shelton's translation of Don Quijote https://spacecruft.org/books/quixote-shelton-ocr
47 commits 1 branch 10 tags 65 KiB
  • Python 100%
Find a file
Exact
Jeff Moe a5509fe2f0 Model requires dummy key. Set docext temp, max tokens
2025-07-08 14:38:55 -06:00
src/shelton Model requires dummy key. Set docext temp, max tokens 2025-07-08 14:38:55 -06:00
.gitignore Ignore output/ dir 2025-07-07 13:44:03 -06:00
CC-1.0-Universal.txt Public domain, CC0 2025-07-07 11:53:50 -06:00
CHANGELOG.txt v0.4.0 2025-07-07 13:55:27 -06:00
LICENSE-apache.txt Apache 2.0 2025-07-07 11:54:02 -06:00
pyproject.toml Add OCR all stub script 2025-07-07 13:36:11 -06:00
README.md Status noted 2025-07-07 13:56:51 -06:00

README.md

OCR of Shelton's Quixote

This repository contains OCR scripts to process Thomas Shelton's translation of Don Quixote.

Status

  • Download: working.
  • Extract: working.
  • Server: working.
  • OCR: working.
  • OCR Everything: working.
  • Compare: TODO.
  • Assemble: TODO.

Install

git clone https://spacecruft.org/books/quixote-shelton-ocr
cd quixote-shelton-ocr/
python -m venv venv
source venv/bin/activate
pip install -U setuptools pip wheel
pip install -e .

For development tools, also do:

pip install -e .[dev]

Scans

The scans come from The Internet Archive. Three copies of the same book are used for QA.

The text is from The Harvard Classics, volume 14.

These files are downloaded, two with JPEG2000, one with TIFF images.

Scripts

Run the scripts in this order:

shelton-download
shelton-extract
shelton-server
shelton-ocr
shelton-compare
shelton-assemble

License

The software is under the Apache 2.0 License.

Unofficial project, not related to upstream projects.

Copyright © 2025 Jeff Moe.

Upstream sources under their respective copyrights.

The resultant OCR text is placed in the public domain.

CC0 1.0 Universal

2025, Jeff Moe