There is a growing interest in the development and deployment of intelligent surveillance systems in public and private locations. This book consists of a coherent selection of extended versions of presentations made in two successful symposia on intelligent distributed surveillance systems (IDSS) and brings together in the latest developments in the field.
Inspec keywords: distributed sensors; sport; video surveillance; object detection; traffic engineering computing; object tracking; video cameras; sensor fusion; closed circuit television
Other keywords: object tracking; uncalibrated arbitrary topology camera networks; distributed database; distributed domestic surveillance system; event detection; CCTV camera; distributed multisensor surveillance system; public transport application; intelligent distributed video surveillance system; football player tracking
Subjects: Image recognition; Other applications of television and video systems; Video signal processing; Closed circuit television; Sensing devices and transducers; Video recording
This chapter discusses state-of-the-art distributed surveillance systems.
In this chapter, the authors show how detailed understandings of everyday surveillance work and organisational conduct may inform the design and development of image processing systems to enhance the awareness and monitoring of complex physical and behavioural environments. The setting in question is the operations rooms of complex interconnecting stations on London Underground and other rapid urban transport systems in Europe. The research presented in this chapter is part of a recent EU-funded project (IST DG VII) known as PRISMATICA, concerned with enhancing support for security management including the development of technologies for surveillance work in the public transport domain. This chapter considers one particular system currently developed to support real-time monitoring of complex environments an integrated surveillance architecture that can process data from multiple cameras and automatically identify particular events. The chapter begins by exploring the tacit knowledge and practices on which operators rely in identifying and managing events and then briefly reflect on the implications of these observations for the design of technical systems to support surveillance. By examining how operators in this domain oversee a complex organisational setting, and the resource on which they rely to identify and manage events, we are considering the everyday practices of monitoring as well as the practical development of the technologies to support it.
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. The database supports the operational and reporting requirements of the surveillance application and is a core component of the quantitative performance evaluation video-tracking framework.
The techniques described in the previous sections have been implemented and inte grated in a system able to access multiple cameras, extract and track people, analyse their behaviour and detect possible dangerous situations, reacting by sending an alarm to a remote PDA with which a video connection is established. The integrated system is structured as a client-server architecture, as shown. The server side contains several pipelined modules: object segmentation, tracking, posture classification and event detection. The alarms generated can be sent to a control centre or can trigger remote connections with a set of authorised users, which exploit the universal multimedia access algorithms to receive visual information of the objects (people) involved in dangerous situations.
This chapter describes a general-purpose system for distributed surveillance and communication.
In this chapter, an approach for recognising targets after occlusion is proposed. It is based on salient reappearance periods discovered from long-term data. By detecting and relating main paths from different regions and using a robust estimate of noise, salient reappearance periods can be detected with high signal-to-noise ratios. Off-line recognition is performed to demonstrate the use of this extracted salient reappearance period and the appearance model to associate and track targets between spatially separated regions. The demonstration is extended to regions between spatially separated views with minimal modifications. As the underlying process of reappearance is not the salient reappearance time but the average distance between paths, the performance of this recognising process is degraded if the average distance between paths is increased. These issues need further investigation.
This chapter presents the distributed camera network designed by INRETS. Four applications have been implemented on this network for the purpose incident detection and statistical monitoring of passenger flows in public transport systems. The authors mainly propose detection algorithms. The domain is to use cameras and computers to perform the desired detection functions.
Sports scenarios present an interesting challenge for visual surveillance applications. Here, we describe a system, and a set of techniques, for tracking players in a football (soccer) environment. The system input is video data from static cameras with overlapping fields-of-view at a football stadium. The output is the real-world, real-time positions of football players during a match. In this chapter, we discuss the problems and solutions that have arisen whilst designing systems and algorithms for this purpose.
Nowadays, the surveillance of important areas such as shopping malls, stations, airports and seaports, is becoming of great interest for security purposes. In the last few years, we have witnessed the installation of a large number of cameras that now are almost covering every place in the major cities. Unfortunately, the majority of such sensors belong to the so-called second generation of CCTV surveillance systems where human operators are required to interpret the events occurring in the scene. Only recently modern and autonomous surveillance systems have been proposed to address the security problems of such areas. Many systems have been proposed for a wide range of security purposes from traffic monitoring to human activity understanding. Video surveillance applications often imply paying attention to a wide area, so different kinds of cameras are generally used, for example, fixed cameras, omni-directional cameras and pan, tilt and zoom (PTZ) cameras.