A text extractor for smart india hackathon
Download the trained model from Google Drive and place it in folder where you want your checkpoint path run
python3 run_demo_server.py --checkpoint_path=/path/to/east_icdar2015_resnet_v1_50_rbox/Instructions for integrating Tessaract with EAST:
install tesseract-ocr
sudo apt-get install tesseract-ocr
sudo apt-get install tesseract-ocr-hin
Get the trained data files, and cube files for reading hindi.
wget https://github.com/indic-ocr/tessdata/blob/master/hin/hin.traineddata
For more trained models, go to https://github.com/indic-ocr/tessdata
Move the downloaded files to /usr/share/tesseract-ocr/tessdata/tessconfigs [Actual location may differ]
To download trained models for other languages, go to https://github.com/indic-ocr/tessdata
sudo mv hin.*cube.* /usr/share/tesseract-ocr/tessdata
sudo mv hin.traineddata /usr/share/tesseract-ocr/tessdata
Create two directories: imcache, and text, in the root directory of this project
mkdir imcache text