#AzureIoT – How to create an Azure IoT module from an Azure #CustomVision project 👀 6/N

Hi !

Today’s post will be mostly focused on code. And on the main tasks

• Add a configuration property to the module to enable / disable the telemetry messages
• Each time the Azure IoT Module process an image, it will trigger a telemetry with the information of the detected image

The sample running show how for each image analyzed (sent using Postman) a new message is triggered to Azure IoT.

Let’s start from here.

Add a property to enable / disable the message trigger

Let’s start adding environment variable to the Azure IoT Module to define if the module will trigger messages.

Then I defined this function to read the environment value

def GetEnvVarBool(name):
  varVal = bool(os.environ[name])
  return varVal

And I reading the env value is as easy as:

trigger_enabled = GetEnvVarBool('ReportValues')

Send an Azure IoT Hub message for each processed image

As I defined in my previous posts, I have a variable named [module_client] that represents the Azure IoT Module. The following function will

• Validate if the trigger messages is enabled
• Load the json response from the Custom Vision image process
• Sort the predictions results by probability
• Add each tagName as a message property
• Trigger the message

async def send_iot_message(strMessage):
    global module_client
    if (trigger_enabled == False):
        return

    message = Message("")

    # load json def and add each tag as properties
    jsonStr = strMessage.get_data(as_text=True)
    jsonObj = json.loads(jsonStr)

    preds = jsonObj['predictions'] 
    sorted_preds = sorted(preds, key=lambda x: x['probability'], reverse=True)
    if (sorted_preds):
        for pred in sorted_preds:
            # tag name and prob * 100
            tagName     = str(pred['tagName'])
            probability = pred['probability'] * 100
            message.custom_properties[tagName] = str(probability)

    await module_client.send_message(message)

And, finally the call to this function is performed on each http endpoint using asyncio.run() to avoid a thread block.

def predict_image_handler(project=None, publishedName=None):
    try:
        imageData = None
        if ('imageData' in request.files):
            imageData = request.files['imageData']
        elif ('imageData' in request.form):
            imageData = request.form['imageData']
        else:
            imageData = io.BytesIO(request.get_data())

        img = Image.open(imageData)
        results = predict_image(img)

        jsonResults = jsonify(results)
        asyncio.run(send_iot_message(jsonResults))

        return jsonResults

So far, so good. I covered a trivial scenario and I think is a good memory backup for me !

Resources

• Azure Custom Vision
• Azure Custom Vision Documentation
• Azure IoT Tools
• Deploy your first IoT Edge module to a virtual Linux device

¿Con ganas de ponerte al día?

En Lemoncode te ofrecemos formación online impartida por profesionales que se baten el cobre en consultoría:

• Si tienes ganas de ponerte al día con Front End (ES6, Typescript, React, Angular, Vuejs…) te recomendamos nuestros Máster Front End: https://lemoncode.net/master-frontend#inicio-banner
• Si te quieres poner al día en Backend (stacks .net y nodejs), te aconsejamos nuestro Bootcamp Backend: https://lemoncode.net/bootcamp-backend#bootcamp-backend/banner
• Y si tienes ganas de meterte con Docker, Kubernetes, CI/CD…, tenemos nuestro Bootcamp Devops: https://lemoncode.net/bootcamp-devops#bootcamp-devops/inicio