VIDEO-BASED PARKING SPACE ANALYSIS FOR TRAFFIC MANAGEMENT

Compared to the analysis of moving traffic, video-based parking space analysis is a less investigated object of research. The aim of this work is to present a new method for the determination of free parking space and the automatic display of improperly parked vehicles with one algorithm under 24-hour conditions. Accordingly, the results will be used in a traffic management system to solve the conflict of objectives between driving time and limited parking space. The specific constraints of the project, in particular predetermined camera height and position, made the development of a new algorithm necessary. The idea is based on edge detection with subsequent Hough transformation. These common techniques are complemented by additional ideas necessary for a 24-hour application. The edge detection operates on an adaptive threshold. The results of the Hough transformation are automatically interpreted. The threshold adaption and the interpretation of the Hough transform are newly developed. The system is built up on the A17 motorway in Germany. All presented results are taken from this real-world case study.