Invoice extractor extract key value pairs from a invoice

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

• Pytesseract

• Flask

• OpenCV

• PILLOW

• Pdf2img

• EAST text detection model

pip install requirements.txt

python app.py

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

• Deployed on Google compute engine - upload your pdf to test

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