In this paper, a video-based system used to collect pedestrian traffic data is presented. It is composed of two different processes. The first process is a motion detection algorithm, which can cope with deformable moving bodies like pedestrians. Its originality consists in the on line construction of a reference image, containing all static edges in the scene. It is particularly robust towards illumination changes. A tracking process is then used to extract pedestrians traffic parameters like occupation rate, and crossing time. This second process is based on a feature matching algorithm and works in real time. The system is adapted to crowd monitoring and used as an intelligent sensor in the project Intelligent Crossroads developed by INRETS (National Research Institute of Transportation and Security). The performance of the video-based system has been evaluated under several real environmental conditions in an outdoor test site. The effects of variations concerning lighting and meteorological conditions (shadows, clouds, sun, rain, night,...) have been carefully tested. Data collected by the system allows control in real time of the traffic in order to improve the service given to pedestrians. This project is supported by INRETS and the Nord Pas de Calais regional council.
Abstract