Advertisement

Shenzhen AI start-up Intellifusion helps city police identify jaywalkers and banned drivers

  • In the first 10 months after installation of the equipment, as many as 13,930 jaywalking offenders were named and shamed

Reading Time:3 minutes
Why you can trust SCMP
Shenzhen is known for its stringent traffic regulations to control the flow of hundreds of thousand of vehicles in the southern Chinese city of 12 million people, with surveillance cameras set up in every corner of the city. Photo: AFP
Li Taoin Shenzhen

The Chinese city of Shenzhen is home to hundreds of thousands of vehicles, so it needs stringent regulations to control the heavy flow of traffic through its streets. But with 12 million residents, enforcing those rules is not easy.

Shenzhen’s police have an invisible helping hand, however, courtesy of local firm Intellifusion, which uses proprietary high-definition cameras and artificial intelligence (AI) technology to catch traffic rules violators, whether they are driving or on foot.

For two years now, Intellivision’s system has been quickly matching car number plates and pedestrians’ faces with information stored in the police system – and displaying the names and faces of jaywalkers on nearby LED screens to shame them into complying.

Advertisement

The current system features a network of cameras, each with a 7 million-pixel resolution, Wang Jun, the company’s director of marketing solutions, told the South China Morning Post in March last year.

In the 10 months after the system’s debut, as many as 13,930 jaywalkers were displayed on the LED screen at one busy intersection in Futian district, according to Shenzhen traffic police.

Intellifusion also plans to team up with the country’s technology giants, such as Tencent, and telecommunications network operators to introduce a system where jaywalkers are notified of their transgressions via instant messaging through smartphone apps such as WeChat – as well as being fined, Wang said in the interview.

Advertisement
Select Voice
Choose your listening speed
Get through articles 2x faster
1.25x
250 WPM
Slow
Average
Fast
1.25x