Typegroups classifier for OCR
Project description
ocrd_typegroups_classifier
Typegroups classifier for OCR
Installation
From PyPI
pip3 install ocrd_typegroup_classifier
From source
If needed, create a virtual environment for Python 3 (it was tested successfully with Python 3.7), activate it, and install ocrd.
virtualenv -p python3 ocrd-venv3
source ocrd-venv3/bin/activate
pip3 install ocrd
Enter in the folder containing the tool:
cd ocrd_typegroups_classifier/
Install the module and its dependencies
make install
Finally, run the test:
sh test/test.sh
** Important: ** The test makes sure that the system does work. For speed reasons, a very small neural network is used and applied only to the top-left corner of the image, therefore the quality of the results will be of poor quality.
Models
The model classifier-1.tgc is based on a ResNet-18, with less neurons per layer than the usual model. It was briefly trained on 12 classes: Adornment, Antiqua, Bastarda, Book covers and other irrelevant data, Empty Pages, Fraktur, Griechisch, Hebräisch, Kursiv, Rotunda, Textura, and Woodcuts - Engravings.
Heatmap Generation
Given a trained model, it is possible to produce heatmaps corresponding to classification results. Syntax:
python3 tools/heatmap.py ocrd_typegroups_classifier/models/classifier.tgc sample.jpg out
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for ocrd_typegroups_classifier-0.0.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 6b339f6b52cb62acc93f64d11637aa895a2cfbe7958df3391e4d6480d8c87d28 |
|
MD5 | 48c202c02d301243c8e9f365e9dcad1d |
|
BLAKE2b-256 | 6ad0620fd50f319ef68ec959b67d0c048bb0f1d602ca5cc0baa0ff46fd235382 |
Hashes for ocrd_typegroups_classifier-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | d469964e37069a2dab403bbf7400eec4ddabcf4ee83c86d6e88bda1bd96e9c1d |
|
MD5 | 19437f8f76a7e346479a2bea163b164f |
|
BLAKE2b-256 | e61b5d0e6967985a7e23d01f558677bd7de4385dacc0186e4896ad23cb4e2f0d |