Cost-effective single-camera multi-car parking monitoring and vacancy detection towards real-world parking statistics and real-time reporting

Abstract

Parking is a huge problem in densely populated areas and drivers spend a significant amount of time finding a suitable place to park their cars. A system that could show drivers the nearest available space would result in enormous savings of time, fuel, and street space. In order to achieve that, real-world periodic statistical analysis of car parking areas could help increase efficiency. Ideally, real-time information could also be used to create personalized suggestions to drivers, thus enabling satisficing of a wide range of possible criteria of optimality. We propose a system that uses a single camera for a wide-area external parking, followed by a combination of two kinds of algorithms: static image analysis of parking lot spaces using a combination of histogram classification and edge detection, and dynamic image analysis using blob analysis. Our system thus achieves monitoring of parking spaces and reports statistics as well as empty slots in real-time. Our results indicate that almost 90% of empty spots are reported correctly, resulting in significant savings through a highly cost-effective single-camera system which can monitor more than 100 spaces. © 2012 Springer-Verlag.