Data and information fusion (DIF) involves a process of combining data from multiple inputs (from one or more sources such as sensors and textual reports).1 The aim is to obtain information that is better (more useful and meaningful) than that would be derived from each of the sources individually (that is, without fusing). DIF is emerging as an important field of multidisciplinary study [113, 316]. This is due to increase in data and information flow, as well as improvement in communication, computing and sensor technology. The first applications of DIF techniques were in the military arena [177, 179, 455]. The use of DIF techniques for problem-solving has extended to many non-military applications in the commercial and industrial sectors [177, 179, 199, 228]. In general, data and information fusion can provide enhancement to the outcomes of processes for solving various application problems. Some advantages of carrying out DIF include [316]: 2 improvement in the accuracy of data, as well as reduction in uncertainty and ambiguity within data, and 2 improvement in situation awareness (SAW) and inference that lead to better decision making.

High-level Information Fusion: An Overview.

: Heterogeneous multisensor systems have been widely used in a variety of military and civilian applications. While the majority of research in multisensor systems is dedicated to military applications, other applications include robot navigation, autonomous vehicles and paramilitary operations. In general, single sensor systems only provide partial information on the state of the environment while multisensor systems rely on data fusion techniques to combine related data from multiple similar and/or dissimilar sensors. The goal of a multisensor system is to provide a synergistic effect that enhances the quality and availability of information about the state of the world over that which would be acquired solely from one sensor. Sensor management can be described as a system or process that provides automatic or semi-automatic control of a suite of sensors or measurement devices. Previous approaches to sensor management all appear to suffer from the mixing of sensor physical requirements with information needs. The result has been ad hoc point solutions that treat the problem as a single optimization task with a performance measure as a weighted sum of diverse, noncommensurate measures. This dissertation presents a new mathematical representation of the multisensor system to capture the sensor management process. Based on this representation, an original hierarchical sensor management model is developed that partitions the system into its constituent processes. These include the sensors themselves, the targets, the Fusion Space, and the Information Space. The Information Space is further partitioned into the Mission Manager, the Information Instantiator, and the Sensor Scheduler. Additionally, this dissertation describes a new approach which uses partially ordered sets to construct a goal-lattice that converts qualitative mission goals to quantitative values for different sensor actions.

A Comprehensive Approach to Sensor Management and Scheduling

Resource Management for Big Data Platforms

This invention relates to an analog video merging system, and more particularly to a system that can take the output of a plurality of video cameras and form a combined video signal which can be fed to a computer having only a single frame grabber card. This system is used to track targets using multiple video cameras. The present invention is comprised of an analog video merging apparatus for merging N video signals from N video cameras. The apparatus is comprised of N video processors, each of the N processors is connected to the N video cameras in a one-to-one correspondence, wherein each of the N video processors compares the video signal associated with one of the N video cameras with a dynamic predetermined threshold level and produces a binary output signal having a low level and a high level, the low level being output if the associated video signal is below the predetermined threshold, and high level being output if the associated video signal is above the predetermined threshold. N binary combiners are provided, with each being connected to the N video processors in a one-to-one correspondence. Each combiner produces a processed binary signal having a low level and a high level, the low level being output in the presence of a low level output from the associated video processor and a high level being output in the presence of a high level from the associated video processor, wherein the high level of the processed binary signal is a unique level to that particular binary combiner such that a unique high level is output from the N binary combiners that are associated with the N video cameras. The processed binary signal for camera N is scaled by a factor of 2N-1 with respect to the other camera signals. A summing unit is provided for summing N processed binary signals from the N binary combiners to produce a merged video signal. A method of providing a merged video signal is also provided.

Analog video merging system for merging N video signals from N video cameras

Information fusion studies theories and methods to effectively combine data from multiple sensors and related information to achieve more specific inferences that could be achieved by using a single, independent sensor. Information fused from sensors and data mining analysis has recently attracted the attention of the research community for real-world applications. In this sense, the deployment of an Intelligent Offshore Oil Industry Environment will help to figure out a risky scenario based on the events occurred in the past related to anomalies and the profile of the current employee (role, location, etc.). In this paper we propose an information fusion model for an intelligent oil environment in which employees are alerted about possible risk situations while their are moving around their working place. The layered architecture, implements a reasoning engine capable of intelligently filtering the context profile of the employee (role, location) for the feature selection of an inter-transaction mining process. Depending on the employee contextual information he will receive intelligent alerts based on the prediction model that use his role and his current location. This model provides the big picture about risk analysis for that employee at that place in that moment.

/pdf/high-level-information-fusion-for-risk-and-accidents-3r8zl82tgo.pdf

High-Level Information Fusion for Risk and Accidents Prevention in Pervasive Oil Industry Environments

The development of a multiple platform distributed fusion capability offers the potential for effective data interchange linking multiple surveillance platforms and centers. Initial off-board derived data can provide the appropriate cues for sector prioritization and sensor resource management. The efficient correlation and fusion of surveillance data from multiple sources with sensors directly on the platform will provide increased situation awareness. This paper discusses a multiplatform distributed fusion architecture under development to provide enhanced cueing and sensor resource management. It shows the current status of ongoing testbed development activities that can be used to provide insight into the proposed architecture. An example is provided to correlate this research to the drug interdiction problem. >

A. Bramson

Papers

Off-board augmented fusion for improved target detection and track