naiiytom / healthdoc-ocr

Health Documents OCR

Description

End-to-end Optical Character Recognition (OCR) project using multiple networks to detect, transform distorted document images, line level segmentation and character recognition.

Demo

docker composer

$ docker-compose up -d

Development

Front End

$ cd frontend
$ docker build -t frontend .
$ docker run -it --rm -p 8080:8080 frontend

Tesseract API

Docker

$ cd tesseract-api
$ docker build -t tesseract-api .
$ docker run -it --rm -p 5000:5000 tesseract-api

References

Tesseract OCR

PyTesseract