Cloud Image Transform

This tool will fetch, re-process, and display images from the web. It uses Pillow for on the fly image manipulation. Run it locally or on a Google Cloud Function.

Quick Example usage

<localhost or Cloud Function>/<your-endpoint>?url=<external-image-url>&w=300&gs=1&fmt=png

Original Image: https://storage.googleapis.com/misc-shared-images-public/rita-pup.png (3.5 MB 1817 x 2837)

Example using ?url=https://storage.googleapis.com/misc-shared-images-public/rita-pup.png&w=300&bg=rgb(9,189,201)&fmt=jpeg&qlt=1

Resize (327 KB, 500 x 780)

Set Background Color (24 KB, 300 x 468)

Reduce File Size (2 KB, 300 x 468)

Parameters:

• url: REQUIRED. url of image to ingest. eg. ?url=https://site.io/image.png
• w: width in pixels. Applying just width will auto scale height. eg. &w=300
• h: height in pixels. Applying just height will auto scale width. eg. &h=150
• fmt: outputs image format in jpeg, png, or webp only. Default is png. eg. &fmt=jpeg
• qlt: for jpeg and webp, decreases the image quality and file size. Default is 90. eg. &qlt=10
• gs: boolean for grayscale. eg. &gs=1
• bg: to set transparency background color. eg. &bg=rgb(9,189,201)
• rotate: Pick from 90, 180, and 270. eg. &rotate=180
• apikey: Check incoming request for validation. eg &apikey=<api-key-you-create>

Components

1. Image Processing Google Cloud Function (image_processor_func/main.py):

   • Fetches an image from a given URL.
   • Optionally resizes the image to specified dimensions.
   • Optionally rotates the image.
   • Optionally converts the image to grayscale.
   • Optionally applies a background color, RGB.
   • Optionally reduces quality of JPEGs and WEBPs.
   • Returns a PNG, JPEG, or WEBP.
   • The entry point for the function is process_image_for_transformation.

2. Local Development Server (local_dev.py):

   • A Flask application that allows local testing of the image processing logic found in image_processor_func.main.process_image_for_transformation.

Setup and Usage

Python Environment

The project uses Conda for environment management. You can create and activate the environment using the environment.yml file:

  conda env create -f environment.yml
  conda activate cloud-image-transform

Create .env File

In root directory of this folder. Update the <values> for Cloud Run and apikey.

  export CLOUD_IMAGE_API_KEY="<api-key-you-create>"
  export GCF_FUNCTION_NAME_ENV_VAR="<your-google-cloud-function-name>"
  export GCF_REGION_ENV_VAR="<GCP-region-you-picked>"

apikey can be optional, just comment out the following in image_processor_func/main.py:

  try:
      api_key = environ.get("CLOUD_IMAGE_API_KEY")
      provided_api_key = flask_request.args.get("apikey")
      if not provided_api_key or provided_api_key != api_key:
          return "Unauthorized: Access is denied due to an invalid API key.", 401
  except Exception as e:
      print(f"Error checking API key: {e}")
      return "Authorization Failed: Access is denied.", 401

This approach uses a header parameter for basic auth, alternatively, security could be achieved directly within GCP's IAM.

Local Development

To test the image processing function locally, run the Flask development server:

python local_dev.py

Google Cloud Function

• Deployment: The Google Cloud Function can be deployed using the deploy script:

  chmod +x deploy.sh
  ./deploy.sh 

• Dependencies: Python dependencies for the cloud function are listed in requirements.txt

Additional Examples

Resize Image

Original Image

Resized Image &w=300

Add Background and reduce quality/file size &bg=rgb(99,200,140)&quality=1(

Make grayscale &gs=1

Rotate! &rotate=90 &rotate=180 &rotate=270

Setting width and height can distort the image &w=300&h=250