Smart Parking at the Pittsburgh International Airport

Finding a good spot in the airport parking space can be a hassle, especially when trying to catch a departing flight. It could be very helpful if one is able to know the location of the nearest empty parking spot. In this project, the authors designed and built a smart airport parking system that achieved this goal. The development of the smart parking system consists of the infrastructure for surveillance cameras, algorithm design for vehicle detection/tracking, and a mobile app for user interface. Collaborating with the Pittsburgh International Airport, 8 surveillance cameras were installed in the long term parking area at the airport. The video streams recorded by these cameras were transmitted to a central server, and analyzed with the vehicle detection and vehicle tracking algorithms. This analysis provided live status updates of the parking spaces available in the observed area. A mobile app was developed which receives the parking space information, specifically the location of best current empty parking spots. The app is also aware of the current position of the user’s vehicle, both through global positioning system (GPS) and the visual tracking of the central tracking server, and provides turn by turn navigation guidance to the user. State-of-the-art object detection and tracking algorithms were applied to obtain the critical information (location of empty spots through vehicle detection and tracking of the user’s vehicle) for the airport smart parking system. To prove the feasibility of these techniques, the authors investigated the performance of these algorithms statistically. They collected a large scale surveillance dataset from the installed camera streams, and conducted experiments to test the accuracy of the vehicle detection and vehicle tracking. The results show that with proper implementation and adaptation, these algorithms are both highly reliable and efficient enough for real-time applications.

Language

  • English

Media Info

  • Media Type: Digital/other
  • Edition: Final Research Report
  • Features: Figures; Photos; References;
  • Pagination: 7p

Subject/Index Terms

Filing Info

  • Accession Number: 01701712
  • Record Type: Publication
  • Report/Paper Numbers: Project 35
  • Contract Numbers: DTRT-13-GUTC-26
  • Files: UTC, TRIS, ATRI, USDOT
  • Created Date: Apr 10 2019 5:18PM