Skip links

Curtin University Smart Parking Project

Curtin University Smart Parking Project

In 2017, Curtin University in Perth, Australia identified a need to find a parking solution to improve business efficiencies and enhance customer satisfaction. After trialling several options including in-ground sensors, Automatic Number Plate Reader (ANPR) count, and various forms of video analytics, Curtin University decided on a camera-based, smart-parking solution to be provided by ParKam.  

ParKam’s solution utilises AI deep machine learning and image processing to capture information regarding parking availability. It allows access to both real-time data as well as historical. What sets it apart from other parking solutions is its ability to analyse an area and not just specific bays, and to identify occupancy levels as well as infringements with 99% accuracy. 

The solution also has the first ever turn-by-turn navigation to a free parking spot using a Google style mapping and linking it to an app. All of these factors make it a highly innovative and unique parking solution.

“Parkam demonstrated a mature, well-developed system that met our needs and exceeded our expectations,” said Graham Arndt Operations & Maintenance Director, Curtin University Perth, “Unlike most other bay-counting technologies, the Parkam system was also able to identify vehicles parked incorrectly, and was able to analyse trends and predict occupancy. It is a very impressive product.”  

Overall, the smart parking solution at Curtin University covered nearly 6000 parking bays across 28 parking lots and was the largest installation of image processing, smart-parking technology in the world!

 

The solution included:

  • 260 cctv cameras installed 
  • 75 brand new poles to be installed throughout the university 
  • 7 new communication external cabinets (CECs) installed  
  • 84 Point-multipoint devices 
  • 47 battery systems (to be powered at night and then run during the day on 24 Volt) 
  • 5 x LED screens on custom made bases

    To minimise disruption to the University’s activities, parameters were put into place including the installation to be completed in 3 months, a timeframe that was easily achieved.

The system is currently undergoing its machine learning phase before being fully operational in early 2020. 

You may also like

Join the Discussion