Videcon has repurposed its deep learning, artificial intelligence software solutions in response to the Government’s guidance for retail premises. Thermi-screen, a mass body temperature screening camera, can pinpoint individuals in a shopping queue or public space who may have an elevated temperature. The device can scan up to 40 people at a time and has a claimed accuracy of ±0.3°C. Additionally, it can also identify anyone not wearing a face covering.
When triggered, the system automatically sends an alert to a designated controller or person in authority. Videcon’s managing director, Matt Rushall, stated, ‘The message that members of the public must wear face coverings is out there, but retailers and managers of public spaces are now finding, as an added burden, they are being asked by the police and authorities to support them in upholding the law. Retailers have a duty to protect their customers, and with police resources stretched to the limit, any technology which helps manage enforcement is to be welcomed.’
Videcon uses facial recognition and mask detection software in its Thermi-scan door access control panels but this is the first time the company has embedded deep learning AI technology in crowd control camera systems.
In addition to the face coverings and mask detection feature, Videcon has enhanced the system’s privacy settings which can pixelate individuals and replace exact temperature readings with simple normal/elevated temperature messaging. This privacy setting secures compliance with GDPR and ensures the maximum level of data protection.
The system comprises of a camera, an NVR and a temperature calibrator. Using two camera modules simultaneously, the system takes temperature readings from the forehead and upper region of the eye orbit. Algorithms compare readings with other environmental factors like room temperature and direct sunlight. The accuracy of Thermi-screen eliminates any false readings, such as a person carrying and sipping a hot drink, and its response time reduces the need for screening of individuals in a line, accelerating the movement of people through a designated screening area.