Skip to content
Cloudbased Image Transformation Service using a IBM Cloudfunction
Python JavaScript HTML CSS
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Cloudbased Image Transformation


This program calls a Cloud Function in the IBM Cloud, wehre it transforms the image given via url into an either
Black and White Image,
a Cartoon Image,
or upscale it by a factor of 2, 3 or 4.

As of now, Github Pages doesn't add a CORS Header to my jQuery ajax POST request. In order to run this application it needs to be run locally, redirecting traffic over the included node.js server

The Cloud Function source code lies in the folder sourceCode Main executable file is


Python 3.7.2+ (If the python code is to be run locally, the libraries Pillow and numpy are needed)
node.js, including the modules express, request and cors
IBM Cloud Account

How to use it

  1. Download the repository
  2. Rename the sourceCode/ to
  3. Upload it as cloud function into your IBM Account either via UI or using the following command via commandline:
    ibmcloud fn action create ACTION_NAME APP_FILE --kind RUNTIME 
    ACTION_NAME: Name you want to give the cloud function 
    APP_FILE: your 
    RUNTIME: Python:3.7 
  4. Go to your IBM Cloud Functions Menu, click Actions, select the Action you created, click Endpoints and Enable as Web Action, now save
  5. Optional:
    Go to API, click Create Managed API, set an API name, add a base path for the API and create an Operation
    PATH: the REST-API path you want it to have, for example: /transform
    Verb: POST
    Package containing action: If your action is in a custom package, select it. Otherwise leave it at (Default)
    Action: Select the Cloudfunctionname you created
    Response content type: application/json
    Click create, then save
  6. Now go to the API Explorer tab, click POST <your Operation PATH>, then copy the Link next to POST.
    Alternatively, if you skipped the API creation, copy the URL of the Web Action we activated in Step 4
  7. Edit the sourceCode/nodejs local/addCORSheader.js you downloaded from this repository and replace the link of the constant apiUrl in line 7 with your link
  8. Run the node.js file via CLI
    node addCORSheader.js
  9. Open the Index.html in your browser
  10. Select the Filter of your choice
  11. add a direct url to a jpeg or png file in the Image URL and, if you used upscale, select a scale modifier
  12. Click Confirm. Note: It'll take a few seconds to load the image

Source of the included Images

Example forrest image by Johannes Plenio, which can be found here
Example image of the 3 Dogs by Helena Lopes, which can be found here

You can’t perform that action at this time.