Jennifer Sander

7 – Bayesian methods for image fusion

The Bayesian fusion methodology bases upon a solid mathematical theory, provides a rich ensemble of methods and allows an intuitive interpretation of the fusion process. It is applicable independently of the goal pursued by image fusion, at different abstraction levels, and also if different kinds of image data have to be fused. It allows transformation, fusion, and focusing, i.e. it fulfils the basic requirements that a reasonable fusion methodology has to satisfy.

Fusion agents - realizing Bayesian fusion via a local approach

Bayesian theory delivers a powerful theoretical platform for the mathematical description and execution of fusion tasks, especially if the information delivering sources are of heterogeneous nature. However, the complexity of Bayesian fusion tasks increases exponentially with the number of sources. A new agent based architecture that is modelled on a successful operating process of the real world, namely criminalistic investigation, circumvents the high computational costs by realizing a local fusion approach. In analogy to criminalists, software agents shall be appointed to perform heterogeneous fusion tasks. In this paper, we give an overview over this concept and the potentials that are provided by it. Then, we focus on translating the proposed ideas into a formal mathematical notation

Using heterogeneous multilevel swarms of UAVs and high-level data fusion to support situation management in surveillance scenarios

The development and usage of Unmanned Aerial Vehicles (UAVs) quickly increased in the last decades, mainly for military purposes. This technology is also now of high interest in non-military contexts like logistics, environmental studies and different areas of civil protection. While the technology for operating a single UAV is rather mature, additional efforts are still necessary for using UAVs in fleets (or swarms). The Aid to SItuation Management based on MUltimodal, MUltiUAVs, MUltilevel acquisition Techniques (ASIMUT) project which is supported by the European Defence Agency (EDA) aims at investigating and demonstrating dedicated surveillance services based on fleets of UAVs. The aim is to enhance the situation awareness of an operator and to decrease his workload by providing support for the detection of threats based on multi-sensor multi-source data fusion. The operator is also supported by the combination of information delivered by the heterogeneous swarms of UAVs and by additional information extracted from intelligence databases. As a result, a distributed surveillance system increasing detection, high-level data fusion capabilities and UAV autonomy is proposed.

A local approach for Bayesian fusion: Mathematical analysis and agent based conception

An agent based architecture that is modelled on a successfully operating process of the real world-criminal investigation-circumvents high computational costs caused by Bayesian fusion by realising a distributed local Bayesian fusion approach. The idea underlying local Bayesian fusion approaches is to perform Bayesian fusion at least not in detail on the whole space that is spanned by the Properties-of-Interest. Local Bayesian fusion is mainly based on coarsening and restriction techniques. Here, we focus on coarsening. We give an overview over the agent based conception and translate the proposed ideas in a formal mathematical framework.

A local approach for focussed Bayesian fusion

Local Bayesian fusion approaches aim to reduce high storage and computational costs of Bayesian fusion which is separated from fixed modeling assumptions. Using the small world formalism, we argue why this proceeding is conform with Bayesian theory. Then, we concentrate on the realization of local Bayesian fusion by focussing the fusion process solely on local regions that are task relevant with a high probability. The resulting local models correspond then to restricted versions of the original one. In a previous publication, we used bounds for the probability of misleading evidence to show the validity of the pre-evaluation of task specific knowledge and prior information which we perform to build local models. In this paper, we prove the validity of this proceeding using information theoretic arguments. For additional efficiency, local Bayesian fusion can be realized in a distributed manner. Here, several local Bayesian fusion tasks are evaluated and unified after the actual fusion process. For the practical realization of distributed local Bayesian fusion, software agents are predestinated. There is a natural analogy between the resulting agent based architecture and criminal investigations in real life. We show how this analogy can be used to improve the efficiency of distributed local Bayesian fusion additionally. Using a landscape model, we present an experimental study of distributed local Bayesian fusion in the field of reconnaissance, which highlights its high potential.

Bayesian fusion: Modeling and application

Bayesian statistics leads to a powerful fusion methodology, especially for the fusion of heterogeneous information sources. If fusion problems are handled under consideration of the full expressiveness and the full range of methods provided by Bayesian statistics, the Bayesian fusion methodology possesses an impressive wide range of applications. We discuss this by having a closer look at selected aspects of Bayesian modeling. Thereby, also parallels to other methods used for information fusion will be drawn. With regard to the practical tractability of Bayesian fusion problems, selected approaches to deal with its potentially high complexity are discussed.

ISR analytics: Architectural and methodic concepts

Prevention and management of damage scenarios require adequate situation awareness to make timely, coordinated, and proactive decisions possible. The stakeholders must be able to access and to comprehend relevant information quickly and with justifiable effort. The resulting challenges for intelligence, surveillance, and reconnaissance (ISR) lie not only in the new and further development of individual sensor and exploitation systems but also in interoperable system networking as well as in the realization of adequate strategies for the collection, processing, dissemination, and presentation of data and information products [1], [2], [3], [4]. In this publication, we present a high level architecture for ISR analytics that complies with these observations. It provides the functionality to customize the system precisely to specific scenarios of the ISR domain. We give a more detailed insight into concepts and approaches that are essential for specific architecture components.

Global evaluation of focussed Bayesian fusion

Information fusion is essential for the retrieval of desired information in a sufficiently precise, complete, and robust manner. The Bayesian approach provides a powerful and mathematically funded framework for information fusion. By local Bayesian fusion approaches, the computational complexity of Bayesian fusion gets drastically reduced. This is done by a concentration of the actual fusion task on its probably most task relevant aspects. In this contribution, further research results on a special local Bayesian fusion technique called focussed Bayesian fusion are reported. At focussed Bayesian fusion, the actual Bayesian fusion task gets completely restricted to the probably most relevant parts of the range of values of the Properties of Interest. The practical usefulness of focussed Bayesian fusion is shown by the use of an example from the field of reconnaissance. Within this example, final decisions are based on local significance considerations and consistency arguments. As shown in previous publications, the absolute values of focussed probability statements represent upper bounds for their global values. Now, lower bounds which are obtained from the knowledge about the construction of the focussed Bayesian model are proven additionally. The usefulness of the resulting probability interval scheme is discussed.

Semantic information fusion to enhance situational awareness in surveillance scenarios

In recent years, the usage of unmanned aircraft systems (UAS) for security-related purposes has increased, ranging from military applications to different areas of civil protection. The deployment of UAS can support security forces in achieving an enhanced situational awareness. However, in order to provide useful input to a situational picture, sensor data provided by UAS has to be integrated with information about the area and objects of interest from other sources. The aim of this study is to design a high-level data fusion component combining probabilistic information processing with logical and probabilistic reasoning, to support human operators in their situational awareness and improving their capabilities for making efficient and effective decisions. To this end, a fusion component based on the ISR (Intelligence, Surveillance and Reconnaissance) Analytics Architecture (ISR-AA) [1] is presented, incorporating an object-oriented world model (OOWM) for information integration, an expressive knowledge model and a reasoning component for detection of critical events. Approaches for translating the information contained in the OOWM into either an ontology for logical reasoning or a Markov logic network for probabilistic reasoning are presented.

Bayes'sche Methodik zur lokalen Fusion heterogener Informationsquellen (Bayesian Methodology for the Local Fusion of Heterogeneous Information Sources)

Bei der Fusion heterogener Informationsquellen muss deren unterschiedlicher Abstraktionsgrad und deren unterschiedliche Natur (Formalisierung) überwunden werden. Essenzielle Forderungen an eine Fusionsmethodik sind die Fähigkeiten zur Transformation, Fusion und Fokussierung. Es wird gezeigt, dass die Bayes’sche Wahrscheinlichkeitstheorie in einer Degree-of-Belief-Deutung jede dieser Forderungen erfüllt. Um ihren hohen Rechenaufwand entscheidend zu verringern, wird anschließend ein lokaler Bayes’scher Fusionsansatz vorgestellt. Dieser kann in Anlehnung an kriminalistische Ermittlungen mittels einer agentenbasierten Fusionsarchitektur umgesetzt werden. In fusing heterogeneous information sources, their different abstraction levels and formalizations have to be coped with. Essential requirements on a fusion methodology are its abilities to transform, fuse, and focus. It is shown that the Bayesian fusion methodology as Degree-of-Belief interpretation covers all these areas. With a view to reduce high computational costs, a local approach for the Bayesian fusion of information is subsequently be presented. In analogy to criminalistic investigation, this approach can be realized via agent-based fusion architecture.