Chinese tech company Hanvon claims to have developed facial recognition technology that can identify people even when they are wearing masks, Reuters reported Monday, as the Covid-19 epidemic has increased the use of masks to avoid potential infections.
The Beijing-based firm told Reuters that it used core technologies developed over the past decade, a sample database of about 6 million unmasked faces and a smaller database of masked faces to develop the facial recognition technology.
About 95% of mask wearers can be recognized by the technology, and the success rate for people without masks is as high as 99.5%, Hanvon vice president Huang Lei was quoted as saying.
Currently, the firm provides two types of products powered by the technology, with one used for face scanning at building entrances and the other more powerful one used for public surveillance camera systems.
Hanvon claims it has about 200 clients in Beijing using its facial recognition technology, including the Ministry of Public Security, which uses the technology to cross-reference images with its own database of names and other information and then detect and track criminal suspects and terrorists. The company also expects to bring the technology to foreign countries as the novel coronavirus spreads globally.
Apart from tech firms, China’s scientific research institutes are also finding solutions to recognizing people donning masks. According to a Hubei Daily report, the National Engineering Research Center for Multimedia Software at Wuhan University has recently come up with a technology that can recognize masked faces with a success rate of nearly 90%. The center has also established what it calls the world’s first open sample database of masked faces.