Self-taught smart car park tech 'could enhance smart cities'

 

Technology consultants have hailed the potential of machine vision and deep learning after unveiling what they say is the world’s smartest car park.

Cambridge Consultants said the system, named Goldeneye, has taught itself to recognise cars and how those cars appear in spaces. It can do this without expensive physical infrastructure and in a range of lighting and weather conditions, day and night, ‘even proving itself during recent severe snow in the UK’.

Goldeneye uses a machine vision and deep learning solution developed at Cambridge Consultants, along with the existing security camera and networking infrastructure on-site, to continuously monitor the availability of parking bays.

While traditional parking monitoring systems use sensors for each parking space, which can be expensive to maintain, Goldeneye using 12 cameras to monitor 430 parking spaces.

Thomas Carmody, head of transport and infrastructure at Cambridge Consultants, said: ‘We’re now at a point where deep learning can move out of the research fields, into the real world, and we’re excited to be pioneering this world first for cities of the future.

‘What’s truly remarkable about Goldeneye is the fact that the system taught itself to identify and operate a car park. It does this without the need for any additional computing equipment.'

The firm said Goldeneye offers a more cost-effective way to scale parking monitoring via the cloud, making it possible to choose your parking location, reserve and pay for a parking spot online while also allowing the user to continually monitor their car. Retailers could harvest valuable information on footfall from vehicles, dwell time at retail locations, car brand and more.

It said that because Goldeneye is based on machine vision it can readily be extended to include new use cases, including occupancy monitoring per vehicle, vehicle identification, and identification of pedestrian flows.

‘Taking the system out of the car park, it could enhance wider smart city applications. Deep learning and machine vision could be harnessed to monitor traffic or to manage crowds on train platforms, for retail analytics, to monitor crowd safety and a range of other applications.’

 
comments powered by Disqus