7.1 Introduction

Public transport operators are facing an increasing demand in efficiency and security from the general public as well as from governments. An important part of the efforts deployed to meet these demands is the ever-increasing use of video surveillance cameras throughout the network, in order to monitor the flow of passengers, enable the staff to be informed of possible congestion and detect incidents without delay. A major inconvenience of this approach, however, is the very large number of cameras required to effectively monitor even a comparatively small network. Added to the cost of the cameras themselves, the cost and complexity of the required wiring, plus the sheer impossibility of watching all the images at the same time, make such a system growingly ineffective as the number of cameras increases.

In recent years, image-processing solutions have been found to automatically detect incidents and make measurements on the video images from the cameras, relieving the staff in the control room of much of the hassle of finding out where interesting events are happening. However, the need remains to bring the latter to a centralised computer located in a technical room, which leaves the need to place huge lengths of video cabling, increasing the cost and the complexity and decreasing the adaptability of the video system.

In the framework of the EU's PRISMATICA programme [ ^[1]], INRETS devised and tested in real-life conditions an architecture to address these problems. The...