Skip to content

Latest commit

 

History

History

03_cv

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
.vscode
.vscode
 
 
client
client
 
 
readme.md
readme.md
 
 

readme.md

Using Azure Computer Vision on a Raspberry Pi

Overview

Let your Raspberry Pi see with Azure Computer Vision! By the end of this tutorial, you'll have your Raspberry Pi detecting objects that are in the frame of the camera. You can try this out or a prototype a product at no cost using a free tier of the computer vision cognitive service.

Prerequisites

  1. An active Azure account. If you don't have one, you can sign up for a free account.
  2. VS Code
  3. Hardware listed below

Hardware

Item Description Link
Raspberry Pi 3 or 4 Single board computer Adafruit
Raspberry Pi Camera V2 Ribbon camera for the Pi Adafruit
USB C power supply (Pi 4) Power cable for Raspberry Pi 4 Adafruit
Micro USB USB power supply (Pi 3) Power cable for Raspberry Pi 3 Adafruit
SD Card with Raspberry Pi OS Operating system for the Pi Adafruit

Setup Azure Resources

Resource Description Link
Computer Vision Computer vision API from Azure Azure

Preparing your environment

  1. If you haven't already, clone this repo to your computer

  2. Open command prompt or terminal and navigate to pi-azure-recipes

  3. In command prompt or terminal type and run code 03_cv. This will open the project folder in VS Code

Create a computer vision resource

  1. First you'll need to create a computer vision resource

  2. Select your subscription

  3. For resource group select Create new, and give your resource group a name

  4. Select a Location that is near you

  5. Give your computer vision resource a name

  6. Select Free F0 for the pricing tier

    Note: You can only have one free tier active per account

  7. After your resource is deployed, select Go to resource

  8. You will need the key and endpoint from this resource to connect your Raspberry Pi to it. Select Keys and Endpoints from the left navigation.

  9. Make a note of the key and endpoint for your resource, you'll use these later on your Pi.

    Rember to treat these like passwords

Setup you Raspberry Pi Device

  1. Connect your Raspberry Pi to a monitor and keyboard or use the the instructions here to setup your pi for SSH

  2. Set up your Pi camera using this guide

  3. Using a USB drive or an SSH file transfer software copy the client folder to the Pi

  4. Install required C libraries

    sudo apt install -y libgfortran5 libatlas3-base

  5. Run the python_environment_setup.sh shell script

  6. Once the script finishes navigate to the client folder press *Ctrl + H to show hidden files.

  7. Open the newly created .env file in a text editor and fill in your key and endpoint

    SUBSCRIPTION_KEY='YOUR-SUBSCRIPTION-KEY'
ENDPOINT='YOUR-ENDPOINT'

  8. Then type

    source .venv/bin/activate
python raspberry_pi_client.py

  9. You should see the picture the camera took and what the Computer Vision service was able to identify in it.

Clean up Resources

If you keep the resources you provisioned you'll continue to incur costs on them. The steps below will walk you through how to clean up your resources.

  1. In the Azure portal navigate to the resource group you created earlier

  2. In the Resource group page, review the list of included resources, and verify that they are the ones you want to delete

  3. Select Delete resource group, and follow the instructions.

Deletion may take a couple of minutes. When it's done, a notification appears for a few seconds. You can also select the bell icon at the top of the page to view the notification.