This page contains a list of abstracts for the peer-reviewed articles at the session on Monday in SWIFT 2009. All abstracts are fetched from camera-ready submissions.
- "Combining multi-sensor fusion and information fusion for increased situation awareness: a research plan" Martin Holmberg and Pontus Svenson
Abstract: In future military operations, the need for interaction between commanders and technical systems at different levels will probably increase due to the changing environment, threats and tactics involved. Since all data/information cannot flow freely in the network for practical and confidentiality reasons, new fusion methods that can handle this situation have to be developed. In this concept paper, we discuss some of the aspects that have to be considered in order to solve this problem.
Keywords: Combining high- and low-level fusion, Information system for commanders at different levels
- "Towards the Implementation of an Ontology-Based Reasoning System for Visual Information Fusion", Juan Gómez-Romero, Jesús García, Miguel A. Patricio and José M. Molina
Abstract: High-level information fusion requires the use of abstract knowledge representation formalisms, such as ontologies. In previous research works, we proposed an ontology-based framework for visual information fusion aimed at contextual recognition of activities and enhancement of basic tracking procedures. In this paper, we describe various aspects of the implementation of this framework, focusing on the development of the supporting rule base.
Keywords: Visual Information Fusion, Ontologies, Reasoning, Context-Awareness
- "Pattern Recognition Using Queries in Relational Tracking Data Bases", Gereon Schüller, Wolfgang Koch and Joachim Biermann
Abstract: Tracking systems provide kinematical information of objects in a scenario. This kinematical information can be combined with additional data to build higher-level-fusion systems that allow the detection of behaviour and threat patterns and thus contribute to situation awareness. The patterns that characterize situations of interest may vary over time and depend on the specific questions to be investigated. Data base systems provide a flexible way of combining data, and standing queries allow ongoing, automatic evaluation. In this paper, we present a way of using data base systems as the central component in a higher-level fusion system. We propose a possible architecture of this system using commercial database management software Finally, we discuss how patterns for the detection of anomalies in tracking scenarios can be expressed in relational algebra.
- "Evolving Petri Net Situation Templates for Situation Recognition", Anders Dahlbom and Lars Niklasson
Abstract: Situation recognition is an important problem to address in order to enhance the capabilities of modern surveillance systems. Situation recognition is concerned with finding a priori defined situations that possibly are instantiated in the present flow of information. It can be a rather tricky task to manually define templates for situations that evolve over time, and to at the same time achieve good results with respect to recall and precision on a situation recognition task. In this paper we present some initial results concerning the task of applying genetic algorithms to evolve Petri net based situation templates of interesting situations. Our results show that it is possible to evolve Petri nets that are on par with manually defined templates. However, more research is needed in order to establish the actual effects it has on recall and precision.
Keywords: Genetic algorithms, Petri nets, situation assessment, situation recognition.
- "Ontological Association in the Ground Target Domain", Jenny Lagerlöf, Egils Sviestins, and Klas Wallenius
Abstract: This paper addresses the problem of associating reports from sensors and human observers to objects in a ground situation picture. Both reports and known objects are assumed to be expressed according to an ontology, in this case expressed in the OWL (Web Ontology Language) formalism. The association here does not consider spatial information; it is entirely based on the class hierarchies and properties. The technique combines results from Information Retrieval Theory for the Semantic Web with an ontology for the Ground Domain and applies to a scenario demonstrating how the potential associations are evaluated and ranked.
Keywords: Ontology, association, ground targets, intelligence
- "Explanation Methods for Bayesian Networks: review and application to a maritime scenario", Tove Helldin and Maria Riveiro
Abstract: Surveillance systems analyze and present vast amounts of heterogeneous sensor data. In order to support operators while monitoring such systems, the identification of anomalous behavior or situations that might need further investigation may reduce operators’ cognitive load. Bayesian networks can be used in order to detect anomalies in data. In order to understand the outcome generated from an anomaly detection application based on Bayesian networks, proper explanations must be given to operators.
This paper presents the findings of a literature analysis regarding what constitutes an explanation, which properties an explanation may have and a review of different explanation methods for Bayesian networks. Moreover, we present the empirical tests conducted with two of these methods in a maritime scenario. Findings from the survey and the experiments show that explanation methods for Bayesian networks can be used in order to provide operators with more detailed information to base their decisions on.
Keywords---anomaly detection, maritime situation awareness, Bayesian networks, explanation methods, Explanation Tree, Causal Explanation Tree.