Filter results by

AWS Rekognition

AWS Rekognition is used for recognizing objects within stored images. It uses AWS Simple Storage Service (S3) as an image file repository.

Once you have the necessary tools installed, you'll take advantage of these services to build a smart trash can.

This tutorial is designed for use with the ARTIK 530 or 710 board with the Ubuntu OS installed.


  • Follow the introductory article to set up the CLI, your account, and your credentials.

  • Set up and test your S3 bucket as a repository for image files.

Using AWS Rekognition

From your ARTIK console, and within the Python 3.5 virtual environment, try to run the following test command. You'll want to copy it to a text editor and change:

  • the bucket name
  • the name of the jpg file to what yours is named in AWS S3
  • the region.

aws rekognition detect-labels --image "{\"S3Object\":{\"Bucket\":\"artiktest\",\"Name\":\"671c9718-bef8-4677-b571-c7e990178c6f.jpg\"}}" --region us-east-1

If everything works, Rekognition will return (as json) the recognition results with confidence numbers. Did it recognize the items in your photo?

Now you're ready to go to the Smart Trash Can article.

Last updated on: