Applicants can now be identified at arrival.

When arriving at the quarantine and training zone the applicants must pass an automatic identity control. Face recognition are used to make sure the identity of the applicants is correct. We have an Rasberry Pi connected to a webcam. Python is used on the Rasberry Pi to control the system. Every 100 ms a picture is taken and uploaded to Azure Face API. 5 pictures of each applicant is already uploaded and available in Azure.

We can then identify the user and will get the correct name of the users. Applicants that are not recognized are not allowed to entry our facilities.

A demonstration of the solution is found here. Notice that the name of the identified applicant is printed in the lower left corner of the screen. The system recognizes Angeliqa in the beginning of the video. Later when Angeliqa moves out of the view of the camera and Simen enters the user shows up as “Unknown” as there is no pictures of Simen uploaded in the Azure image base. Simen would thus not be allowed to enter our facilities.

https://drive.google.com/open?id=1Hp91AuyT9G1-kOK2bWcrbuQoY4Q9LoLE

Badge

Having used the Azure Cloud (Azure Face API), with a webcamera taking pictures of the user we think we deserve the [Cloud city] and [Eyes wide shutter] badges.

And by delivering this before lunch we also think we deserve the [First brunch] badge.