Wearing face masks at public places has become a norm today. Several governments even made it mandatory for people to wear face masks to curb the spread of the pandemic. But, bringing such a large scale behavioral change in a short duration of time is almost impossible. Manually identifying people at public places who are not wearing face masks and persuading them to wear it, is not a practical solution. This is where we need to rely on technology. Today, AI-powered video analytics solutions can identify whether a person in a crowd is wearing a face mask or not. These solutions can be used in public places such as airports, subway stations, restaurants, etc to detect people without facemasks and notify the nearest security officer to intervene and take necessary actions. However, even this solution is not scalable for worldwide adoption.
The majority of face mask detection solutions available in the market today require custom hardware and CCTV infrastructure which cannot be set up by all organizations immediately. This is when we thought about building a face mask detection system that can run in ARM-based devices and embedded devices.
ARM processors are extensively used in consumer electronic devices such as smartphones, tablets, multimedia players, and other mobile devices, such as wearables. Even if a restaurant or a store doesn’t have sophisticated CCTV infrastructure, they would still have basic embedded systems or smartphones. What if an automated face-mask detection system can be built on such devices? We think it can enable a larger number of organizations, businesses, public places to deploy automated face mask detection systems to detect people who are not wearing face masks and notify the business owner or security officials to take necessary actions. Here is a demo video of what we built,
Such a system can enable the majority of businesses, organizations to set up face mask detection systems even if they do not have a sophisticated CCTV infrastructure integrated with real-time video analytics.