Document processing experiments using traditional and advance OCR libraries.
You'll need >= Python 12 to run the notebooks in this repo.
The Jupyter notebooks guide you step-by-step experimentation to use Tesseract and PaddleOCR on the sample documents to see the limitations and improvement suggestions for information extraction.
- Please clone the repository and install all the dependencies to run the notebooks.
https://github.com/atarora/doc-processing-advance-ocr
- It is advised to create a virtual env and install dependencies as :
$ cd doc-processing-advance-ocr
$ python -m venv {venv_name}
$ source {venv_name}/bin/activate
(venv_name) $ pip install -r requirements.txt
More here : https://realpython.com/python-virtual-environments-a-primer/#create-it
- Run jupyter and open notebooks as :
$ python jupyter