ifantrifandi / go-tesseract-ocr-service

Golang based microservice for OCR powered by Tesseract.

go-tesseract-ocr-service

This Golang based project provides a microservice that offers a REST API and a Web view to convert PDF's and Images to Text, using Tesseract OCR scanner.

Just a proof-of-concept at this point. For future development it will be split in a multi-tier application architecture for better escalability - again for instructional purposes.

---

1. How to build and run:

docker build -t ocr-tesseract .

docker run --privileged=true -d -t -i \
    -p 8080:80 \
    -e UPLOADED_FILES_DIR='/tmp/pdf-cache' \
    -v /tmp/pdf-cache:/tmp/pdf-cache ocr-tesseract

2. Main Web Views

The service provides some minimalistic webviews to use the functionalities.

http://localhost:8080/web/pdf
http://localhost:8080/web/img

3. Endpoints

3.1 API Endpoints for PDF submission

http://localhost:8080/api/upload/pdf

3.2 API endpoint for Image submission

http://localhost:8080/api/upload/img

4. Frameworks

This projects uses the following SDK's:

• Tesseract OCR : OCR Engine
• GhostScript: PDF interpreter used to convert PDF to a set of images (per page)

(C) 2019