ClaireCut Server

Run

2. Download project

    $ git clone https://github.com/clairecut/clairecut-backend.git
    $ cd clairecut-backend
   

3. This project uses docker, so start the container by running

    $ docker-compose up
   

4. If you see a { hello: world } response to a GET request to $ <YOUR_DOCKER_IP>/api/v1 the server is running.

5. Further test the back-end with a POST request to

    $ <YOUR_DOCKER_IP>/api/v1/submit
   

   and include a base64 encoded image as payload with parameter image.

Test

Test with one word in the image:

    $ python test0.py

Test with six words on the image

    $python test1.py

Future Tasks

1. Validation

   • Received string is base64-image string
   • Add JSON web token authentication
     • New \authenticate endpoint

2. Create clean docker image and upload

3. Make algorithm robust

   • Detect whether image is suitable or not. (imageTropical.json gives several results for blue, but it is not a whiteboard)
   • Did not work with .gif image (do we need that?)

further tasks: see Issues of this repo.

Tipps

To undestand the image processing algorithm, look at processor/README.md

Delete .pyc files with:

find . | grep -E "(__pycache__|\.pyc|\.pyo$)" | xargs rm -rf