This survey aims to provide a comprehensive status of recent and current research on context-based Information Fusion (IF) systems, tracing back the roots of the original thinking behind the development of the concept of ''context''. It shows how its fortune in the distributed computing world eventually per-meated in the world of IF, discussing the current strategies and techniques, and hinting possible future trends. IF processes can represent context at different levels (structural and physical constraints of the scenario, a priori known operational rules between entities and environment, dynamic relationships modelled to interpret the system output, etc.). In addition to the survey, several novel context exploita-tion dynamics and architectural aspects peculiar to the fusion domain are presented and discussed.

In this paper we present a trajectory clustering method suited for video surveillance and monitoring systems. The clusters are dynamic and built in real-time as the trajectory data is acquired, without the need of an off-line processing step. We show how the obtained clusters can be successfully used both to give proper feedback to the low-level tracking system and to collect valuable information for the high-level event analysis modules.

No abstract

Abstract-Moving toward the implementation of the intelligent building idea in the framework of ambient intelligence, a video security application for people detection, tracking, and counting in indoor environments is presented in this paper. In addition to security purposes, the system may be employed to estimate the number of accesses in public buildings, as well as the preferred followed routes. Computer vision techniques are used to analyze and process video streams acquired from multiple video cameras. Image segmentation is performed to detect moving regions and to calculate the number of people in the scene. Testing was performed on indoor video sequences with different illumination conditions.

The concepts of event and anomaly are important building blocks for developing a situational picture of the observed environment. We here relate these concepts to the JDL fusion model and demonstrate the power of Markov Logic Networks (MLNs) for encoding uncertain knowledge and compute inferences according to observed evidence. MLNs combine the expressive power of first-order logic and the probabilistic uncertainty management of Markov networks. \ud Within this framework, different types of knowledge (e.g. a priori, contextual) with associated uncertainty can be fused together for situation assessment by expressing unobservable complex events as a logical combination of simpler evidences. \ud We also develop a mechanism to evaluate the level of completion of complex events and show how, along with event probability, it could provide additional useful information to the operator.\ud Examples are demonstrated on two maritime scenarios of rules for event and anomaly detection

T he importance of video surveillance techniques [3], [9], [25] has increased considerably since the latest terrorist incidents. Safety and security have become critical in many public areas, and there is a specific need to enable human operators to remotely monitor activity across large environments such as: a) transport systems (railway transportation, airports, urban and motorway road networks, and maritime transportation), b) banks, shopping malls, car parks, and public buildings, c) industrial environments, and d) government establishments (military bases, prisons, strategic infrastructures, radar centers, and hospitals).Modern video-based surveillance systems (see classifications described in [9]) employ real-time image analysis techniques for efficient image transmission, color image analysis, event-based attention focusing, and model-based sequence understanding. Moreover, cheaper and faster computing hardware combined with efficient and versatile sensors create complex system architectures; this is a contributing factor to the increasingly widespread deployment of multicamera systems.These multicamera systems can provide surveillance coverage across a wide area, ensuring object visibility over a large range of depths. They can also be employed to disambiguate occlusions. Techniques that address handover between cameras (in configurations with shared or disjoint views) are therefore becoming increasingly more important. Events of interest (identified as moving objects and people) must be then coordinated in the multiview system, and events deemed of special interest must be tracked throughout the scene (see Figure 1). Wherever possible, tracked events should be classified and their dynamics (sometimes called behavior) analyzed to alert an operator or authority of a potential danger. For security awareness based on multiscale spatio-temporal tracking, see Hampampur et al. [10].In the development of advanced visual-based surveillance systems, a number of key issues critical to successful operation must be addressed. The necessity of working with complex scenes characterized by high variability requires the use of specific and sophisticated algorithms for video acquisition, camera calibration, noise filtering, and motion detection that are able to learn and adapt to changing scene, lighting, and weather conditions. Working with scenes characterized by poor structure requires the use of robust pattern recognition and statistical methods. The use of clusters of fixed cameras, usually grouped in areas of interest but also scattered across the entire scene, requires automatic methods of compensating for chromatic range differences, synchronization of acquired data (for overlapping and nonoverlapping views), estimation of correspondences between and among overlapping views, and registration with local Cartesian reference frames.This article describes the low-level image and video processing techniques needed to implement a modern visual-based surveillance system. In particular, change detection methods for both ...

No abstract

No abstract