Serverless Image Recognition v2⚡🔎

A simple API for the images recognition using AWS Rekognition on the back-end. Using the Serverless framework for describing infrastructure.

Overview:

A simple API for the recognition of images using AWS Rekognition on the back-end. The API stores an image, does image recognition and labeling on it and returns results to the user in two ways, with a provided callback or a using GET /blobs/{blob_id} endpoint.

This is an updated version from image-recognition-api:

• Applying code best practice.
• Applying better project structure.
• Applying better logging and error handling.

Architecture diagram

---

Usage:

• Clone the repository:

  $ git clone https://github.com/kfrawee/serverless-image-recognition.git

• Navigate to the directory:

  $ cd .\serverless-image-recognition\

This project uses serverless framework ⚡. So, make sure you get that first and give the necessary permissions to serverless cli. Follow this page for getting started.

• Deploy the stack:

  $ serverless deploy
  
  ✔ Service deployed to stack image-recognition-dev

• After deployment is successful, you can check the deployed stack details using:

  $ serverless info
  
  service: image-recognition
  stage: dev
  region: us-east-1
  stack: image-recognition-dev
  endpoints:
  POST - https://***.execute-api.us-east-1.amazonaws.com/dev/blobs
  GET - https://***.execute-api.us-east-1.amazonaws.com/dev/blobs/{blob_id}
  functions:
  createBlob: image-recognition-dev-createBlob
  getBlob: image-recognition-dev-getBlob
  labelImage: image-recognition-dev-labelImage
  callback: image-recognition-dev-callback
  layers:
  pyimage: arn:aws:lambda:us-east-1:***:layer:pyimage:42

Workflow:

1- Send a POST request to Create blob endpoint POST /blobs to get a blob_id and a pre-signed url to upload your image:

• You can use cURL to send requests or you can use Postman or REQBIN.
• You can use webhook.site to get a callback url to receive status updates.

• Request:
  • cURL:

    curl --location --request POST 'https://***.execute-api.us-east-1.amazonaws.com/dev/blobs' \
    --header 'Content-Type: application/json' \
    --data-raw '{
        "callback_url": "https://webhook.site/41d5eaf6-3213-4dfe-863d-0f3436dc461c"
    }'

  • Postman:
    

    
    Create blob

• Response:

  {
      "blob_id": "01GEJ202KZ7PEBZ36X401352AE",
      "invocation_status": "ACTION_REQUIRED",
      "requested_on": "2022-10-04 17:51:46.815566+00:00",
      "upload_url": "https://s3.amazonaws.com/image-recognition-bucket-dev/blobs/01GEJ202KZ7PEBZ36X401352AE/01GEJ202KZ7PEBZ36X401352AE?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=...",
      "callback_url": "https://webhook.site/41d5eaf6-3213-4dfe-863d-0f3436dc461c",
      "_links": {
          "status": "https://siw4d3y44i.execute-api.us-east-1.amazonaws.com/dev/blobs/01GEJ202KZ7PEBZ36X401352AE"
      }
  }

  • Use the upload_url to upload your image.
  • Use the blob_id to retrieve status from GET /blobs/{blob_id}
  • Also, you will recieve status update on the callback_url - if provided.

2- Upload your image:

• cURL:

  curl --location --request PUT 'https://s3.amazonaws.com/image-recognition-bucket-dev/blobs/01GEJ202KZ7PEBZ36X401352AE/01GEJ202KZ7PEBZ36X401352AE?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=...' \
  --header 'Content-Type: image/jpeg' \
  --data-binary '@/PATH/TO/YOUR/IMAGE/test1.jpeg'

• Postman:
  

  
  Upload image

3- Retrieve Status:

• Request:
  • cURL:

    curl --location --request GET 'https://siw4d3y44i.execute-api.us-east-1.amazonaws.com/dev/blobs/01GEJ202KZ7PEBZ36X401352AE'

  • Postman:
    

    
    Retrieve status

Response:

{
    "blob_id": "01GEJ202KZ7PEBZ36X401352AE",
    "invocation_status": "COMPLETED",
    "requested_on": "2022-10-04 17:51:46.815566+00:00",
    "started_on": "2022-10-04 17:53:14.793551+00:00",
    "completed_on": "2022-10-04 17:53:15.236018+00:00",
    "labels": [
        "Car",
        "Sports Car",
        "Coupe",
        "Tire",
        "Wheel"
    ],
    "callback_url": "https://webhook.site/41d5eaf6-3213-4dfe-863d-0f3436dc461c"
}

Cleaning:

To remove the stack and all the services and resources:

$ serverless remove

✔ Service image-recognition has been successfully removed