Decision Fusion Systems Research Articles

Decision Fusion Using Fuzzy Logic Type 1 for Two Aviation Scenarios

In this paper we discuss some aspects of decision fusion using fuzzy logic of type 1 for two aviation scenarios. We study the fuzzy inference system/engine and briefly discuss the MATLAB-GUI based tool developed by the authors to evaluate any existing or new fuzzy implication functions. Two examples: i) formation flight, and ii) flying along an air lane are numerically evaluated and the performance of decision fusion systems incorporating fuzzy logic is illustrated. The results indicate that some existing as well as new implication functions, devised by the authors, work very well for the examples considered. This opens up new possibilities of applications of fuzzy logic based decision fusion to varieties of aerospace/aviation problems in data fusion, decision making, and decision fusion.

Optimum consultation for serial distributed detection systems

<p>This paper considers a distributed detection system which consists of sensors that are connected in series. The observations of each sensor in this system design are considered to be statistically independent of all other sensors. In contrast to the popular serial decision fusion systems, we assume that consultations are allowed in a serial manner between successive sensors that make up the system. In addition, the paper demonstrates the similarity between the proposed consulting serial system and the optimal serial one in terms of detection probabilities for a give probability of false alarm. However, it should be emphasized that the proposed system has the benefit of conditional nonrandom consultation among the sensors. Consequently, its survivability is higher than that of serial systems. Numerical evaluations for the cases of two and three sensors are provided and compared with those of the serial as well as the centralized schemes.</p>

Deep feature learning versus shallow feature learning systems for joint use of airborne thermal hyperspectral and visible remote sensing data

ABSTRACT Recently, the development of various remote sensing sensors has provided more reliable information and data for identification of different ground classes. Accordingly, multisensory fusion techniques are applied to enhance the process of information extraction from complementary airborne and spaceborne remote sensing data. Most of previous research in the literature has focused on the extraction of shallow features from a specific sensor and on classification of the resulted feature space using decision fusion systems. In recent years, Deep Learning (DL) algorithms have drawn a lot of attention in the machine learning area and have had different remote sensing applications, especially on data fusion. This study presents two different feature-learning strategies for the fusion of hyperspectral thermal infrared (HTIR) and visible remote sensing data. First, a Deep Convolutional Neural Network (DCNN)-Support Vector Machine (SVM) was utilized on the features of two datasets to provide the class labels. To validate the results with other learning strategies, a shallow feature model was used, as well. This model was based on feature fusion and decision fusion that classified and fused the two datasets. A co-registered thermal infrared hyperspectral (HTIR) and Fine Resolution Visible (Vis) RGB imagery was available from Quebec of Canada to examine the effectiveness of the proposed method. Experimental results showed that, except for the computational time, the proposed deep learning model outperformed shallow feature-based strategies in the classification performance that was based on its accuracy.

Dynamic Fuzzy Rule-based Source Selection in Distributed Decision Fusion Systems

ABSTRACTA key challenge in decision fusion systems is to determine the best performing combination of local decision makers. This selection process can be performed statically at the training phase or dynamically at the execution phase, taking into consideration various features of the data being processed. Dynamic algorithms for the selection of competent sources are generally more accurate, but they are also computationally more intensive and require more memory. In this research, we propose a fuzzy rule-based approach for dynamic source selection (FDSS) that compresses the knowledge from local sources using a divide-and-conquer strategy along with the basic concepts of coverage and truth value criteria, leading to less memory requirement and faster processing. A top-down approach to FDSS is then used to reach a parameter-free algorithm, i.e. one that avoids the restrictive parameters/threshold settings of FDSS. The rule bases in both approaches are created recursively and use the conditional probabilities of each class's correctness as the rule's weight. The proposed approaches are compared against several competing dynamic classifier selection methods based on local accuracy. Results indicate that the proposed fuzzy rule structures are generally faster and require less memory, while they also lead to more accurate decisions from the uncertain decisions from multiple sources.

Decision fusion systems for fault detection and identification in industrial processes

Numerous fault detection and identification methods have been developed in recent years, whereas, each method works under its own assumption, which means a method works well in one condition may not provide a satisfactory performance in another condition. In this paper, we intend to design a fusion system by combining results of various methods. To increase the diversity among different methods, the resampling strategy is introduced as a data preprocessing step. A total of six conventionally used methods are selected for building the fusion system in this paper. Decisions generated from different models are combined together through the Dempster-Shafer evidence theory. Furthermore, to improve the computational efficiency and reliability of the fusion system, a new diversity measurement index named correlation coefficient is defined for model pruning in the fusion system. Fault detection and identification performances of the decision fusion system are evaluated through the Tennessee Eastman process.

An iterative method for decision fusion in multiple sensor systems

Distributed decision fusion has been intensively studied in the past. This interest has been sparked by the requirement of a distributed surveillance system to be more reliable and immune to electronic attack than a single sensor system. In this paper, we consider a binary decentralized distributed decision fusion in which a system of multiple independent sensors monitors a common volume and provides relevant binary decisions about the state of the environment to a data fusion center. The fusion center combines the binary decisions of the individual distributed sensors into a final global decision. We propose a simple iterative method for optimizing multiple sensor decision fusion systems, in terms of both the sensors and the fusion center. The proposed iterative method determines, for a given global false alarm probability, the corresponding optimum setting of the individual sensors thresholds and the fusion center combining strategy that yields maximum global detection probability. The proposed method enables efficient search for the optimum solution by starting from a variety of initial trial values. The performance of the proposed method is provided in case of Rayleigh distributed observations and is proved to be cost effective and efficient.