3.1 Introduction

Image surveillance and monitoring is an area being actively investigated by the machine vision research community. With several government agencies investing significant funds into closed circuit television (CCTV) technology, methods are required to simplify the management of the enormous volume of information generated by these systems. CCTV technology has become commonplace in society to combat anti-social behaviour and reduce other crime. With the increase in processor speeds and reduced hardware costs it has become feasible to deploy large networks of CCTV cameras to monitor surveillance regions. However, even with these technological advances there is still the problem of how information in such a surveillance network can be effectively managed. CCTV networks are normally monitored by a number of human operators located in a control room containing a bank of screens streaming live video from each camera.

This chapter describes a system for visual surveillance for outdoor environments using an intelligent multi-camera network. Each intelligent camera uses robust techniques for detecting and tracking moving objects. The system architecture supports the real-time capture and storage of object track information into a surveillance database. The tracking data stored in the surveillance database is analysed in order to learn semantic scene models, which describe entry zones, exit zones, links between cameras, and the major routes in each camera view. These models provide a robust framework for coordinating the tracking of objects between overlapping and non-overlapping cameras, and recording the activity of objects detected by the system.