Creation of a Mask Detection model running on the edge with Azure AI Dev Kit

1. Creation of a custom vision model using Microsoft customvision.ai (https://www.customvision.ai)
2. Saving the AI Dev Kit custom vision model into an Azure blob Storage.
3. Adding the location of the custom vision model into the Azure Edge device configuration.
4. Then you can use our video camera with the new custom vision model running directly on the EDGE in order to detect mask or no mask attributes from a video stream.

Presentation

https://github.com/retkowsky/MaskDetection/blob/master/Mask%20Detection%20with%20Azure%20AI.pdf

Microsoft Custom Vision.ai interface

https://www.customvision.ai/

More information about AI Dev Kit

https://azure.github.io/Vision-AI-DevKit-Pages/

AI@Edge community pages

https://microsoft.github.io/ai-at-edge/

Additional demo

https://github.com/Azure-Samples/IoTDemos/tree/master/WorkplaceHealthAndSafetyDemo

More information about Azure ML for Data Science

https://docs.microsoft.com/en-us/azure/machine-learning/

Azure IoT Hub

https://docs.microsoft.com/en-us/azure/iot-hub/

Intelligent Edge device

https://azure.microsoft.com/en-us/overview/future-of-cloud/

Thanks to Laurent Maymard - IoT GBB (laurent.maymard@microsoft.com) - for his Azure IOT & Edge expertise.

Serge Retkowsky
serge.retkowsky@microsoft.com
https://www.linkedin.com/in/serger/
https://www.youtube.com/channel/UCiqfuuwDAFJGtOw_LTAtbFg