Queue Prediction with Alerting
We tell you when a queue is going to form, minutes before it actually forms. By understanding the flow of customers, Queue Alert sends predictive alerts to store staff to open or close tills to minimise queuing time and optimise staff resources.
Unlike most vision-based counting systems, we use a camera that observes a certain store area from a horizontal plane of 10–30°. This setup allows using the video stream for many other purposes such as passer-by traffic, footfall, demographics (age and gender), journey map, dwell times or promotional analysis.
Prediction on customer distribution can help shop managers to plan ahead and set the optimal staff-to-customer ratio. The web-based platform records and analyses the data gathered from video feeds and create a proposed staff schedule for the upcoming periods with a click of a button.
Privacy and data protection
The platform can anonymise (blur faces and/or full bodies) video streams real-time. An individual’s identity is protected by converting faces into mathematical expressions that are encrypted and cannot be reversed. By re-identification of a person, only these vectors are matched, individuals are not identified.