Vision-based intelligent control of transportation systems

In response to the overcrowding of transportation systems, there has been an increase in the emphasis place upon Intelligent Vehicle/Highway Systems (IVHS). Whether the application is keeping a safe distance between vehicles on highways or allowing a pedestrian to safely cross a street, environmental sensing and control play a crucial role in IVHS applications. In this paper we present robust visual detection and tracking techniques for IVHS applications, where computer vision may be the only sensor used or a part of a larger, multi-sensor system. We have selected the application of pedestrian detection and tracking to demonstrate that the Controlled Active Vision framework can provide precisely the type of information required to effectively manage traffic situations.