Decision Fusion Approach Research Articles

TAAWUN: a Decision Fusion and Feature Specific Road Detection Approach for Connected Autonomous Vehicles

Road transportation is among the global grand challenges affecting human lives, health, society, and economy, caused due to road accidents, traffic congestion, and other transportation deficiencies. Autonomous vehicles (AVs) are set to address major transportation challenges including safety, efficiency, reliability, sustainability, and personalization. The foremost challenge for AVs is to perceive their environments in real-time with the highest possible certainty. Relatedly, connected vehicles (CVs) have been another major driver of innovation in transportation. In this paper, we bring autonomous and connected vehicles together and propose TAAWUN, a novel approach based on the fusion of data from multiple vehicles. The aim herein is to share the information between multiple vehicles about their environments, enhance the information available to the vehicles, and make better decisions regarding the perception of their environments. TAWUN shares, among the vehicles, visual data acquired from cameras installed on individual vehicles, as well as the perceived information about the driving environments. The environment is perceived using deep learning, random forest (RF), and C5.0 classifiers. A key aspect of the TAAWUN approach is that it uses problem specific feature sets to enhance the prediction accuracy in challenging environments such as problematic shadows, extreme sunlight, and mirage. TAAWUN has been evaluated using multiple metrics, accuracy, sensitivity, specificity, and area-under-the-curve (AUC). It performs consistently better than the base schemes. Directions for future work to extend the tool are provided. This is the first work where visual information and decision fusion are used in CAVs to enhance environment perception for autonomous driving.

Multimodal Gait Recognition With Inertial Sensor Data and Video Using Evolutionary Algorithm

Evolutionary decision fusion has applications in biometric authentication and verification. Gray wolf optimizer (GWO) is one such evolutionary decision fusion approach that can be used to tune the fusion parameters in a multimodal data acquisition system. Human gait is a proven biometric trait with applications in security and authentication. However, acquiring human-gait data can be erroneous due to various factors and multimodal fusion of such erroneous gait data can be challenging. In this paper, we propose a new decision fusion-based approach to solve the above problem. Gait data is recorded simultaneously using motion sensors and visible-light camera. The signals of the motion sensors are modeled using a long short-term memory neural network and corresponding video recordings are processed using a three-dimensional convolutional neural network. GWO has been used to optimize the parameters during fusion. It has been chosen based on the underlying hunting strategy that leads to better approximation of the solution. Interestingly, in our case it converges quicker than other optimization techniques such as genetic algorithm or particle swarm optimization. To test the model, a dataset involving 23 males and females has been recorded while they perform four different types of walks, including, normal walk, fast walk, walking while listening to music , and walking while watching multimedia content on a mobile . An overall accuracy of 91.3% has been recorded across all test scenarios. Results reveal that the proposed study can further be explored to design robust gait biometric systems.

Using Text Mining and Data Mining Techniques for Applied Learning Assessment

In a society where first hand work experience is greatly valued many universities or institutions of higher education have designed their Quality enhancement plan (QEP) to address student applied learning. This paper is the results of a university’s QEP plan, called Experiencing Transformative Education Through Applied Learning or ETEAL. This paper will highlight the research that was conducted using text mining and data mining techniques to analyze a dataset of 672 student evaluations collected from 40 different applied learning courses from fall 2013 to spring 2015, in order to evaluate the impact on instructional practice and student learning. Text mining techniques are applied through the NVivo text mining software to find the 100 most frequent terms to create a document-term matrix in Excel. Then, the document-term matrix is merged with the manual interpretation scores received to create the applied learning assessment data. Lastly, data mining techniques are applied to evaluate the performance, including Random Forest, K-nearest neighbors, Support Vector Machines (with linear and radial kernel), and 5-fold cross-validation. Our results show that the proposed text mining and data mining approach can provide prediction rates of around 67% to 85%, while the decision fusion approach can provide an improvement of 69% to 86%. Our study demonstrates that automatic quantitative analysis of student evaluations can be an effective approach to applied learning assessment.

Machine Learning Algorithms and Fault Detection for Improved Belief Function Based Decision Fusion in Wireless Sensor Networks.

Decision fusion is used to fuse classification results and improve the classification accuracy in order to reduce the consumption of energy and bandwidth demand for data transmission. The decentralized classification fusion problem was the reason to use the belief function-based decision fusion approach in Wireless Sensor Networks (WSNs). With the consideration of improving the belief function fusion approach, we have proposed four classification techniques, namely Enhanced K-Nearest Neighbor (EKNN), Enhanced Extreme Learning Machine (EELM), Enhanced Support Vector Machine (ESVM), and Enhanced Recurrent Extreme Learning Machine (ERELM). In addition, WSNs are prone to errors and faults because of their different software, hardware failures, and their deployment in diverse fields. Because of these challenges, efficient fault detection methods must be used to detect faults in a WSN in a timely manner. We have induced four types of faults: offset fault, gain fault, stuck-at fault, and out of bounds fault, and used enhanced classification methods to solve the sensor failure issues. Experimental results show that ERELM gave the first best result for the improvement of the belief function fusion approach. The other three proposed techniques ESVM, EELM, and EKNN provided the second, third, and fourth best results, respectively. The proposed enhanced classifiers are used for fault detection and are evaluated using three performance metrics, i.e., Detection Accuracy (DA), True Positive Rate (TPR), and Error Rate (ER). Simulations show that the proposed methods outperform the existing techniques and give better results for the belief function and fault detection in WSNs.

Data Augmentation in Deep Learning-Based Fusion of Depth and Inertial Sensing for Action Recognition

This article covers a deep learning-based decision fusion approach for action or gesture recognition via simultaneous utilization of a depth camera and a wearable inertial sensor. The deep learning approach involves using a convolutional neural network (CNN) for depth images captured by a depth camera and a combination of CNN and long short–term memory network for inertial signals captured by a wearable inertial sensor, followed by a decision-level fusion. Due to the limited size of the training data, a data augmentation procedure is carried out by generating depth images corresponding to different orientations of the depth camera and by generating inertial signals corresponding to different orientations of the inertial sensor placement on the body. The results obtained indicate the positive impact of the decision-level fusion as well as the data augmentation on the recognition accuracies.

Fusion of multispectral imagery and DSMs for building change detection using belief functions and reliabilities

ABSTRACTThe extraction of building changes from very high resolution satellite images is an important but challenging task in remote sensing. Digital Surface Models (DSMs) generated from stereo imagery have proved to be valuable additional data sources for this task. In order to efficiently use the change information from the DSMs and spectral images, belief functions have .been introduced. In this article, two-step building change detection fusion models based on both Dempster-Shafer Theory (DST) and Dezert-Smarandache Theory (DSmT) frameworks are proposed. In the first step, basic belief assignments (BBAs) of the change indicators from images and DSMs are calculated by using a refined sigmoidal BBA model. Then these BBAs are employed for the new proposed building change detection decision fusion approach. In order to cover the miss-detections introduced by the wrong height values of the DSMs and incomplete information from images, disparity maps from the DSM generation procedure and shadow maps from the multispectral channels are adopted to generate reliability maps, which are further integrated to the fusion models. In the last step, building change masks are generated based on four decision-making criteria. In the experimental part of this work, we evaluate the performance of this new building change detection method on real satellite images thanks to a building change reference mask representing the ground truth. Substantial accuracy improvements are achieved when comparing the new results with those obtained from classical 3D change detection approaches.

A pervasive electroencephalography-based person authentication system for cloud environment

Cloud security is a major concern to the current research community. Existing solutions are vulnerable to various threats and can easily be forged. Electroencephalography (EEG) signals can provide unique identification that is being explored by various research groups for the development of biometric solutions. In this paper, we present a person authentication framework suitable for cloud environment using EEG signals. EEG signals are recorded in a mobile device while the participants listen to music. Recorded signals are then transferred to a cloud server using Representational State Transfer (REST) web service, where useful features are extracted. Person identification and verification processes are done using two well known classifiers, namely Hidden Markov Model (HMM) and Support Vector Machine (SVM). We have improved the performance of the system using a decision fusion approach. A total of 40 volunteers have participated in this study for collecting brain activity data. Identification accuracy has been recorded as 97.5%. Effectiveness of the proposed framework has been validated using Receiver Operating Characteristic (ROC) curve with 100% True Positive Rate (TPR) and 35% False Acceptance Rate (FAR).

Hybrid of discrete wavelet transform and adaptive neuro fuzzy inference system for overall driving behavior recognition

Monitoring and evaluating of driving behavior is the main goal of this paper that encourage us to develop a new system based on Inertial Measurement Unit (IMU) sensors of smartphones. In this system, a hybrid of Discrete Wavelet Transformation (DWT) and Adaptive Neuro Fuzzy Inference System (ANFIS) is used to recognize overall driving behaviors. The behaviors are classified into the safe, the semi-aggressive, and the aggressive classes that are adopted with Driver Anger Scale (DAS) self-reported questionnaire results. The proposed system extracts four features from IMU sensors in the forms of time series. They are decomposed by DWT in two levels and their energies are sent to six ANFISs. Each ANFIS models the different perception about driving behavior under uncertain knowledge and returns the similarity or dissimilarity between driving behaviors. The results of these six ANFISs are combined by three different decision fusion approaches. Results show that Coiflet-2 is the most suitable mother wavelet for driving behavior analysis. In addition, the proposed system recognizes the overall driving behavior patterns with 92% accuracy without necessity to evaluate the maneuvers one by one. We show that without longitude acceleration data, the driver behavior cannot be recognized successfully while the results do not disturb substantially when the gyroscope is not available.

A generic framework for decision fusion in Fault Detection and Diagnosis

In this paper, we propose a unified framework that enables decisions fusion for applications dealing with multiple heterogeneous Fault Detection and Diagnosis (FDD) methods. This framework, which is a discrete Bayesian Network (BN), is generic and can encompass all FDD method, whether it requires an accurate model or historical data. The main issue concerns the integration of different decisions emanating from individual FDD methods in order to obtain more reliable results.The methodology is based on a theoretical learning of the BN parameters, according to the FDD objectives to be reached. The development leads to a multi-objective problem under constraints, which is solved with a lexicographic approach.The effectiveness of the proposed decision fusion approach is validated through the Tennessee Eastman Process (TEP), which represents a challenging industrial benchmark. The application demonstrates the viability of the approach and highlights its ability to ensure a significant improvement in FDD performances, by providing a high fault detection rate, a small false alarm rate and an effective strategy for the resolution of conflicts among different FDD methods.

Deep learning decision fusion for the classification of urban remote sensing data

Multisensor data fusion is one of the most common and popular remote sensing data classification topics by considering a robust and complete description about the objects of interest. Furthermore, deep feature extraction has recently attracted significant interest and has become a hot research topic in the geoscience and remote sensing research community. A deep learning decision fusion approach is presented to perform multisensor urban remote sensing data classification. After deep features are extracted by utilizing joint spectral–spatial information, a soft-decision made classifier is applied to train high-level feature representations and to fine-tune the deep learning framework. Next, a decision-level fusion classifies objects of interest by the joint use of sensors. Finally, a context-aware object-based postprocessing is used to enhance the classification results. A series of comparative experiments are conducted on the widely used dataset of 2014 IEEE GRSS data fusion contest. The obtained results illustrate the considerable advantages of the proposed deep learning decision fusion over the traditional classifiers.

Mapping Cropland Abandonment in the Aral Sea Basin with MODIS Time Series

Cropland abandonment is globally widespread and has strong repercussions for regional food security and the environment. Statistics suggest that one of the hotspots of abandoned cropland is located in the drylands of the Aral Sea Basin (ASB), which covers parts of post-Soviet Central Asia, Afghanistan and Iran. To date, the exact spatial and temporal extents of abandoned cropland remain unclear, which hampers land-use planning. Abandoned land is a potentially valuable resource for alternative land uses. Here, we mapped the abandoned cropland in the drylands of the ASB with a time series of the Normalized Difference Vegetation Index (NDVI) from the Moderate Resolution Imaging Spectroradiometer (MODIS) from 2003–2016. To overcome the restricted ability of a single classifier to accurately map land-use classes across large areas and agro-environmental gradients, “stratum-specific” classifiers were calibrated and classification results were fused based on a locally weighted decision fusion approach. Next, the agro-ecological suitability of abandoned cropland areas was evaluated. The stratum-specific classification approach yielded an overall accuracy of 0.879, which was significantly more accurate ( p < 0.05) than a “global” classification without stratification, which had an accuracy of 0.811. In 2016, the classification results showed that 13% (1.15 Mha) of the observed irrigated cropland in the ASB was idle (abandoned). Cropland abandonment occurred mostly in the Amudarya and Syrdarya downstream regions and was associated with degraded land and areas prone to water stress. Despite the almost twofold population growth and increasing food demand in the ASB area from 1990 to 2016, abandoned cropland was also located in areas with high suitability for farming. The map of abandoned cropland areas provides a novel basis for assessing the causes leading to abandoned cropland in the ASB. This contributes to assessing the suitability of abandoned cropland for food or bioenergy production, carbon storage, or assessing the environmental trade-offs and social constraints of recultivation.

Decision fusion in cognitive radio using improved fuzzy approach

Increasing demand for services such as mobile telephones and many others has introduced a need for a fresh perspective on spectrum utilization and management. A promising technology that enhances the utilization of available radio spectrum resources is the cognitive radio technology. Spectrum sensing is the process of monitoring a specific frequency band periodically, to identify the presence or absence of primary users and thus forms the key technology in cognitive radio. Hard and soft decision rules are used to perform data fusion and a final decision is made. Fuzzy based approaches are proved to be effective in decision fusion, but the computational complexity of the same is high. In this work, we have proposed a medium rule reduction technique to optimize the performance of a fuzzy based decision fusion approach. Performance analysis of the this approach is also carried out for various number of neighborhoods. It is seen that the computational complexity of the proposed method is able to reduce the computational complexity with acceptable level of performance.

Energy-Efficiency Joint Cooperative Spectrum Sensing and Power Allocation Scheme for Green Cognitive Radio Network: A Soft Decision Fusion Approach

This paper focus on a jointly spectrum sensing parameter and energy efficiency (EE) optimization problem in OFDMA CRN system for enabling Green Communication. In this perspective, we firstly propose an algorithm to choose less spatially-correlated cognitive users to reduce the shadowing effect in wireless network. Furthermore, based on Lagrangian duality theorem with the aid of parametric transformation, the algorithm called an Iterative Dinkelbach Scheme (IDS) is proposed to optimize both transmission power allocation and sensing duration of the cognitive users (Cus) for maximizing Energy Efficiency under the constraints of overall outage of cognitive network, interference to the PU, maximum transmission power and minimum data rate requirement. Numerical result proves the effectiveness of our proposed algorithm. Compared with existing schemes, our proposed scheme outperforms in enhancing the EE with the same parameters.

EEG based word familiarity using features and frequency bands combination

The appearance of unknown words often disturbs communication and reading. This paper focuses on detecting those words which are unfamiliar to the users while reading using Electroencephalography (EEG). In particular, a word-familiarity prediction approach based on EEG signals from the user’s brain waves has been developed. Word-familiarity refers whether a user is familiar with the word or not while reading the text. The recorded signals have been processed using Wavelet decomposition technique and four features, namely mean, standard deviation, Root Mean Square (RMS) and power have been computed from beta and gamma frequency bands. The prediction of word-familiarity has been performed using Random Forest (RF) classifier. Next, a decision fusion approach has also been used to boost the prediction performance. A dictionary based pop-up window has been developed to provide the meaning of the word when a user is found to be unfamiliar with the text. EEG dataset of 12 users has been developed while they are reading 25 words. The results show that the characteristics of brain waves at the time of unknown word perception can be detected, i.e., when the user comes across an unknown word then the pattern of his brain waves comes out to be totally different from the pattern when he is familiar with the word. Similarly, the pattern of brain waves has been found to be different for the word for which the user has confusion in his mind. An accuracy of 82% has been recorded using the proposed classifier combination approach. Finally, a comparative study with other popular classification technique is also discussed.

SHADOWDETECTION FROM VHR AERIAL IMAGES IN URBAN AREA BY USING 3D CITY MODELS AND A DECISION FUSION APPROACH

Abstract. In VHR(very high resolution) aerial images, shadows indicating height information are valuable for validating or detecting changes on an existing 3D city model. In the paper, we propose a novel and full automatic approach for shadow detection from VHR images. Instead of automatic thresholding, the supervised machine learning approach is expected with better performance on shadow detection, but it requires to obtain training samples manually. The shadow image reconstructed from an existing 3D city model can provide free training samples with large variety. However, as the 3D model is often not accuracy, incomplete and outdated, a small portion of training samples are mislabeled. The erosion morphology is provided to remove boundary pixels which have high mislabeling possibility from the reconstructed image. Moreover, the quadratic discriminant analysis (QDA) which is resistant to the mislabeling is chosen. Further, two feature domains, RGB and ratio of the hue over the intensity, are analyzed to have complementary effects on better detecting different objects. Finally, a decision fusion approach is proposed to combine the results wisely from preliminary classifications from two feature domains. The fuzzy membership is a confidence measurement and determines the way of making decision, in the meanwhile the memberships are weighted by an entropy measurements to indicate their certainties. The experimental results on two cities in the Netherlands demonstrate that the proposed approach outperforms the two separate classifiers and two stacked-vector fusion approaches.

A hybrid MLP-CNN classifier for very fine resolution remotely sensed image classification

The contextual-based convolutional neural network (CNN) with deep architecture and pixel-based multilayer perceptron (MLP) with shallow structure are well-recognized neural network algorithms, representing the state-of-the-art deep learning method and the classical non-parametric machine learning approach, respectively. The two algorithms, which have very different behaviours, were integrated in a concise and effective way using a rule-based decision fusion approach for the classification of very fine spatial resolution (VFSR) remotely sensed imagery. The decision fusion rules, designed primarily based on the classification confidence of the CNN, reflect the generally complementary patterns of the individual classifiers. In consequence, the proposed ensemble classifier MLP-CNN harvests the complementary results acquired from the CNN based on deep spatial feature representation and from the MLP based on spectral discrimination. Meanwhile, limitations of the CNN due to the adoption of convolutional filters such as the uncertainty in object boundary partition and loss of useful fine spatial resolution detail were compensated. The effectiveness of the ensemble MLP-CNN classifier was tested in both urban and rural areas using aerial photography together with an additional satellite sensor dataset. The MLP-CNN classifier achieved promising performance, consistently outperforming the pixel-based MLP, spectral and textural-based MLP, and the contextual-based CNN in terms of classification accuracy. This research paves the way to effectively address the complicated problem of VFSR image classification.

A message passing approach for decision fusion in adversarial multi-sensor networks

We consider a simple, yet widely studied, set-up in which a Fusion Center (FC) is asked to make a binary decision about a sequence of system states by relying on the possibly corrupted decisions provided by byzantine nodes, i.e. nodes which deliberately alter the result of the local decision to induce an error at the fusion center. When independent states are considered, the optimum fusion rule over a batch of observations has already been derived, however its complexity prevents its use in conjunction with large observation windows.In this paper, we propose a near-optimal algorithm based on message passing that greatly reduces the computational burden of the optimum fusion rule. In addition, the proposed algorithm retains very good performance also in the case of dependent system states. By first focusing on the case of small observation windows, we use numerical simulations to show that the proposed scheme introduces a negligible increase of the decision error probability compared to the optimum fusion rule. We then analyse the performance of the new scheme when the FC makes its decision by relying on long observation windows. We do so by considering both the case of independent and Markovian system states and show that the obtained performance are superior to those obtained with prior suboptimal schemes. As an additional result, we confirm the previous finding that, in some cases, it is preferable for the byzantine nodes to minimise the mutual information between the sequence system states and the reports submitted to the FC, rather than always flipping the local decision.

Fuzzy decision fusion approach for loss-given-default modeling

In this paper, fuzzy decision fusion techniques are applied to predict loss-given-default of corporate bonds. In our model, we add the principal components derived from more than 100 macroeconomic variables as explanatory variables. However, in order to improve the performance of the model, the Box–Cox transformation of macroeconomic variables is applied prior to loss-given-default modeling. A differential evolution algorithm is used to create an optimized fuzzy rule-based model that fuses the outputs of several base models. We compare the predictions from fuzzy decision fusion techniques with support vector regression techniques, regression trees and OLS regressions. Our findings show that fuzzy decision fusion techniques increase prediction accuracy of loss-given-default modeling and transformations of macroeconomic factors do not affect prediction accuracy of fuzzy models.

Multi-modal video event recognition based on association rules and decision fusion

In this paper, we propose a multi-modal event recognition framework based on the integration of feature fusion, deep learning, scene classification and decision fusion. Frames, shots, and scenes are identified through the video decomposition process. Events are modeled utilizing features of and relations between the physical video parts. Event modeling is achieved through visual concept learning, scene segmentation and association rule mining. Visual concept learning is employed to reveal the semantic gap between the visual content and the textual descriptors of the events. Association rules are discovered by a specialized association rule mining algorithm where the proposed strategy integrates temporality into the rule discovery process. In addition to frames, shots and scenes, the concept of scene segment is proposed to define and extract elements of association rules. Various feature sources such as audio, motion, keypoint descriptors, temporal occurrence characteristics and fully connected layer outputs of CNN model are combined into the feature fusion. The proposed decision fusion approach employs logistic regression to formulate the relation between dependent variable (event type) and independent variables (classifiers’ outputs) in terms of decision weights. Multi-modal fusion-based scene classifiers are employed in the event recognition. Rule-based event modeling and multi-modal fusion capability are shown to be promising approaches for event recognition. The decision fusion results are promising and the proposed algorithm is open to the fusion of new sources for further improvements. The proposal is also open to new event type integrations. The accuracy of the proposed methodology is evaluated on the CCV and Hollywood2 dataset for event recognition and results are compared with the benchmark implementations in the literature.

Decision fusion for hyperspectral image classification based on multiple features and locality-preserving analysis

ABSTRACTA novel fusion-classification system is proposed for hyperspectral image classification. Firstly, spectral derivatives are used to capture salient spectral features for different land-cover classes and a Gabor filter is applied to extract useful spatial features at neighbouring locations. Then, two locality-preserving dimensionality reduction methods are employed to reduce the dimensionality of data and preserve the local structure of neighbouring samples in the original image, derivative-feature and Gabor-feature domains. Finally, the classification results from Gaussian-mixture-model classifiers are fused by a decision-fusion approach. We have compared the proposed system to several traditional and state-of-the-art methods on two benchmark classification data sets. In both cases, our system achieved improved accuracy than the current best performing methods. Especially in the case of classification for the Indian Pines data set, the proposed DG-locality-preserving nonnegative matrix factorization (LPNMF) has 7% higher accuracy than support vector machine-Markov random field and LPNMF-G-MRF, and also has about 4% higher accuracy than Gabor-LPNMF. The proposed method has practical relevance as it shows potential to create smooth classification maps even for complex, spatially heterogeneous land-cover classes.