Fast recognition and reidentification of vehicles from video data for different viewing directions

Abstract

Traffic data can be obtained using video cameras. The data can be used, e.g., to optimise traffic flow by controlling traffic lights dynamically. In order to minimise the number of cameras it us useful to reidentify vehicles leaving ne monitored region and afterwards entering the viewing field of a further camera. From reidentified vehicles travel times can be obtained which are relevant parameters to optimise traffic control. A method to reidentify vehicles based on extraction of 3D prototype vehicle models and on colour extraction from the top plne of vehicles is presented. Shadows and light refelctions on wet street are corrected, and therefore, the high recognition accuracy is achieved which is necessary to find the top plane of the vehicles. The algorithms are siutable for real-time applications.