Showing papers in "Information Fusion in 2017"

TL;DR: This first of its kind, comprehensive literature review of the diverse field of affective computing focuses mainly on the use of audio, visual and text information for multimodal affect analysis, and outlines existing methods for fusing information from different modalities.

TL;DR: It is concluded that although various image fusion methods have been proposed, there still exist several future directions in different image fusion applications and the researches in the image fusion field are still expected to significantly grow in the coming years.

TL;DR: A new multi-focus image fusion method is primarily proposed, aiming to learn a direct mapping between source images and focus map, using a deep convolutional neural network trained by high-quality image patches and their blurred versions to encode the mapping.

TL;DR: This paper surveys research on ensembles for data stream classification as well as regression tasks and discusses advanced learning concepts such as imbalanced data streams, novelty detection, active and semi-supervised learning, complex data representations and structured outputs.

TL;DR: This survey discusses clear motivations and advantages of multi-sensor data fusion and particularly focuses on physical activity recognition, aiming at providing a systematic categorization and common comparison framework of the literature, by identifying distinctive properties and parameters affecting data fusion design choices at different levels.

TL;DR: This overview reviews theoretical underpinnings of multi-view learning and attempts to identify promising venues and point out some specific challenges which can hopefully promote further research in this rapidly developing field.

TL;DR: This paper introduces general NLP techniques which are required for text preprocessing, and investigates the approaches of opinion mining for different levels and situations, and introduces comparative opinion mining and deep learning approaches for opinion mining.

TL;DR: In this article, a personalized individual semantics (PIS) model is proposed to personalize individual semantics by means of an interval numerical scale and the 2-tuple linguistic model, and a new CW framework is defined, such a CW framework allows us to deal with PIS to facilitate CW keeping the idea that words mean different things to different people.

TL;DR: The DHHFL-MULTIMOORA method is applied to deal with a practical case about selecting the optimal city in China by evaluating the implementation status of haze controlling measures and some comparisons are provided to show the advantages of the proposed method.

TL;DR: A method based on the sentiment analysis technique and the intuitionistic fuzzy set theory to rank the products through online reviews and decision support system can be developed to support the consumers purchase decisions more conveniently.

TL;DR: The advances of DFE algorithms for networked systems are reviewed, including data quantization, random transmission delays, packet dropouts, fading measurements and communication disturbances, and some random phenomena induced by networks are discussed.

TL;DR: This paper presents a review of techniques used to gain information about atmospheric dispersion events using static or mobile sensors and discusses on the current limitations of the state of the art and recommendations for future research.

TL;DR: This paper proposes a novel boundary finding based multi-focus image fusion algorithm, in which the task of detecting the focused regions is treated as finding the boundaries between the focused and defocused regions from the source images.

TL;DR: Based on the syntax and semantics of virtual linguistic terms, VLTs could be a possible alternative for solving some current challenges of qualitative information fusion in decision making.

TL;DR: The probabilistic hesitant fuzzy set (P-HFS) has been put forward, it adds the probability to the HFS and can retain more information than the H FS, but it is still not perfect.

TL;DR: A time-varying filter such that, in the presence of the multiple missing measurements, event-triggered transmission mechanism and stochastic nonlinearities, an upper bound of the filtering error covariance is obtained and then minimized by properly designing the filter gain.

TL;DR: This work considers the use of these types of orthopair fuzzy sets as a basis for the system of approximate reasoning introduced by Zadeh, referred to as OPAR, and looks at the formulation of the ideas of possibility and certainty using these orthop air fuzzy sets.

TL;DR: Several systems and architectures related to the combination of biometric systems, both unimodal and multimodal, are overviews, classifying them according to a given taxonomy, and a case study for the experimental evaluation of methods for biometric fusion at score level is presented.

TL;DR: This paper proposes a dead-reckoning (DR)/WiFi fingerprinting/magnetic matching (MM) integration structure that uses off-the-shelf sensors in consumer portable devices and existing WiFi infrastructures and reduces the rate of mismatches by over 75.0% when compared with previous DR/WiFi/MM integration structures.

TL;DR: The extent by which one may depart from this classical centralized paradigm is explored, looking at decentralized anomaly detection based on unsupervised machine learning, to detect anomalies at the sensor nodes, as opposed to centrally, to reduce energy and spectrum consumption.

TL;DR: A journey through the main information fusion ingredients that a recipe for the design of a CBIR system should include to meet the demanding needs of users is offered.

TL;DR: A new augmentation approach is proposed by which the multi-rate sampled-data system under consideration is transformed into the single-rate system and a new fusion estimation scheme with the help of covariance intersection (CI) method is proposed.

TL;DR: This study extends the state of the art of deep learning convolutional neural network (CNN) to the classification of video images of echocardiography, aiming at assisting clinicians in diagnosis of heart diseases.

TL;DR: The extent to which the ensemble implementation improves the overall performance of the selection process, in terms of predictive accuracy and stability, is evaluated and the impact of the ensemble approach on the final selection outcome is measured, i.e. on the composition of the selected feature subsets.

TL;DR: This paper develops fusion rules based on generalizations of the well-known Locally-Optimum Detection (LOD) framework based on Generalized Likelihood Ratio Test (GLRT), Bayesian and hybrid approaches, aimed at reducing the computational complexity.

TL;DR: The purpose of this paper is to design a set of time-varying state estimators such that the dynamics of the state estimation error satisfies the average H ∞ performance constraints.

TL;DR: The ensemble learning algorithm (LSBoost or Bagging), similar to the neural network, can build the SINS/GPS position model based on current and some past samples of SINS velocity, attitude and IMU output information.

TL;DR: Experimental results show that the proposed TransM-RKELM (Transfer learning mixed and reduced kernel Extreme Learning Machine) is a fast and robust human activity recognition model that can adapt the classifier to new sensor locations quickly and obtain good recognition performance.

TL;DR: This paper addresses the distributed fusion filtering problem for discrete-time random signals from measured outputs perturbed by random parameter matrices and correlated additive noises from sensor networks with transmission random packet dropouts.

TL;DR: This paper proposes the use of non-parametric statistical analysis for comparisons of fusion algorithms along with the Image fusion Toolbox Employing Significance Testing (ImTEST).