/flask-ocr

Primary LanguagePythonMIT LicenseMIT

Inovoice extractor

Invoice extractor extract key value pairs from a invoice

Getting Started

Here we have used the East text detection model to detect the texts. after detecting the texts it will go through the tesseract and extract the key values

Prerequisites

  • Pytesseract

  • Flask

  • OpenCV

  • PILLOW

  • Pdf2img

  • EAST text detection model

Installing

pip install requirements.txt
python app.py

Running the tests

The project is showing only 20% accuracy on test set. but it'can be improved by making a custom model for text detection,language support and tweaking the image preprocessing

Deployment

License

This project is licensed under the MIT License - see the LICENSE.md file for details