Text Renderer
Generate text images for training deep learning ocr model.
Setup
The code was only tested on Ubuntu 16.04.
Install dependencies:
pip3 install -r requirements.txt
Generate image
Run python3 main.py
, images and labels.txt will generate at output/default/
.
Run python3 main.py --help
to see optional arguments.
Strict mode
If some chars in corpus is not supported by your font, your will get bad result:
Run main.py
with --strict
, renderer will retry get sample from corpus until all chars are supported by a font.
Tools
Check how many chars your font not support for a charset:
python3 tools/check_font.py
checking font ./data/fonts/eng/Hack-Regular.ttf
chars not supported(4971):
['第', '朱', '广', '沪', '联', '自', '治', '县', '驼', '身', '进', '行', '纳', '税', '防', '火', '墙', '掏', '心', '内', '容', '万', '警','钟', '上', '了', '解'...]
0 fonts support all chars(5071) in ./data/chars/chn.txt:
[]
Debug mode
Run python3 main.py --debug
will save images with extract information.
You can see how perspectiveTransform works and all bounding/rotated boxes.
Generate image using GPU
If you want to use GPU to speed up image generating, first compile opencv with CUDA. Compiling OpenCV with CUDA support
Then build Cython part, and add --gpu
options when run main.py
cd libs/gpu
python3 setup.py build_ext --inplace