Validation of Automatic Passenger Counting in Public Transportation Systems

The counting of vehicles and the monitoring and simulation of their movements in traffic situations for purposes of planning and controlling traffic and traffic environments has been an important research subject for many years. Since roughly one decade, this research subject has been extended to models and studies for passenger and pedestrian traffic. One important issue is the automatic reliable counting of persons, i.e. pedestrians, visitors, passengers, in various traffic situations. Counting of persons appears to be a simple operation, however, automatic counting is subject to a large number of technical problems and to various kinds of errors. Usually the automatic counting of persons is carried out by means of optical sensors or video recording. The counting results heavily depend on the technical properties of the sensors, on the environment in which measuring is done and on many situation dependent factors. The main problem consists in the need for a validation of the counting process. The growing quality requirements have led to the development of 3d techniques in this context. In this paper we sketch the main concepts of an applied research project in which we investigate concepts and software solutions for various sensor environments to address this problem. The main and most obvious application will be the counting of passengers in railway trains, subway trains, busses and tramways. Passenger counting has become an important issue in the management and organization of large public passenger transportation systems. The present work can be considered as a continuation of the results presented at IMS 2007 [4], [5].