Video surveillance cameras that recognise your face, detect your gender, the colour of your clothes and how fast you’re travelling, and then track your movements, will be up and running in Perth by the end of the month.
The City of Perth will switch on the Briefcam technology in new surveillance cameras in East Perth as part of its $1 million Smart Cities trial. The technology, which also includes heat-mapping, will initially be tested in three of the 30 cameras in the new network, and potentially rolled out to more later.
City of Perth commissioner Andrew Hammond said the facial recognition technology would compare the faces of people captured on live footage with photographs supplied by law enforcement officials.
“The facial recognition capabilities will only be activated at the request of relevant law enforcement agencies,” he said.
The City of Perth will have the ability to move the detection and analytical capabilities between all the cameras in the network.
The Briefcam system uses “deep learning”, when a computer develops neural networks trained to recognise patterns, and “computer vision”, when a computer is trained to interpret and understand the visual world.
The software, which has been developed in Israel, will allow the City to be alerted if someone enters a restricted area defined by borders on the screen or if an object has been in the same spot for a certain time. It can also count pedestrians on footpaths and vehicles on roads.
The City of Perth is believed to have joined Border Force, which uses the technology at its passport gates at Perth Airport, as the only WA authority to use facial recognition surveillance on the public.
Chinese company Huawei is aggressively investing in the tech race, spending $21.2 billion on research and development last year. Hong-Eng Koh, Huawei’s global chief of public safety, said facial recognition alone was not sufficient because people could cover their face, so the company was supplying technology that does even more.
“Now we are meeting companies around the world, not just China, but in Europe, US, Israel that have different analytics, what colour you are wearing, how you are walking, what bag you are carrying, or your vehicle,” he said in a briefing in Shenzhen, China, last month.
How it works:
1: The Briefcam system can understand an entire scene and its background.
2: It detects, tracks, extracts and classifies every object or person.
3: Moving objects are separated from the video background, using artificial intelligence, and then tracked.
4: Objects are classified, for example people and vehicles, their attributes e.g. gender, carrying a bag, vehicle model, and saved in a database with their locations.
5: The information in the database can be searched, generate reports and send alerts.
6: The system allows hours of footage to be reviewed in minutes by simultaneously displaying events that have occurred at different times.
This content was originally published here.