Showing papers in "Information Fusion in 2019"

TL;DR: This paper proposes a novel method to fuse two types of information using a generative adversarial network, termed as FusionGAN, which establishes an adversarial game between a generator and a discriminator, where the generator aims to generate a fused image with major infrared intensities together with additional visible gradients.

TL;DR: A novel method for multi-sensor data fusion based on a new belief divergence measure of evidences and the belief entropy was proposed, which outperforms other related methods where the basic belief assignment of the true target is 89.73%.

TL;DR: This survey presents various ML-based algorithms for WSNs with their advantages, drawbacks, and parameters effecting the network lifetime, covering the period from 2014–March 2018.

TL;DR: The so-called “smartization” of manufacturing industries has been conceived as the fourth industrial revolution or Industry 4.0, a paradigm shift propelled by the upsurge and progressive maturity of the global economy.

TL;DR: This work provides the reader with the basic concepts necessary to build an ensemble for feature selection, as well as reviewing the up-to-date advances and commenting on the future trends that are still to be faced.

TL;DR: Experimental results confirm the effectiveness of the proposed system involving the CNNs and the ELMs, which is evaluated using two audio–visual emotional databases, one of which is Big Data.

TL;DR: The focus of this review is to provide in-depth and comprehensive analysis of data fusion and multiple classifier systems techniques for human activity recognition with emphasis on mobile and wearable devices.

TL;DR: The experiments on a public multimodal physiological signal dataset show that the DBN, and FGSVM based model significantly increases the accuracy of emotion recognition rate as compared to the existing state-of-the-art emotion classification techniques.

TL;DR: In this paper, the authors describe the principles of data integration and discuss current methods and available implementations, as well as current challenges in biomedical integrative methods and their perspective on the future development of the field.

TL;DR: An intelligent detection system that is based on Genetic Algorithm and Random Weight Network is proposed to deal with email spam detection tasks and can automatically identify the most relevant features of the spam emails.

TL;DR: The performance of the different categories of pansharpening methods developed between 2000 and 2016 is evaluated based on the idea of meta-analysis, by making a statistical analysis of the studies ever published.

TL;DR: The adaptive multi-view issues for further research in the area of feature selection and fusion are presented by learning view-specific weights to each view data automatically.

TL;DR: In this article, a multi-perspectives classification of the data fusion to evaluate the smart city applications is presented, where the proposed classification is applied to evaluate selected applications in each domain of smart city and the potential future direction and challenges of data fusion integration are discussed.

TL;DR: The properties of IoTData, a number of IoT data fusion requirements including the ones about security and privacy, classify the IoT applications into several domains and a thorough review on the state-of-the-art of data fusion in main IoT application domains are investigated.

TL;DR: An overview of MULTIMOORA is conducted by categorizing and analyzing main researches, theoretically and practically, in terms of the subordinate ranking methods, ranking aggregation tools, weighting methods, group decision-making, combination with other models, and the robustness of the method.

TL;DR: A first hand classification of region based fusion methods is carried out and a comprehensive list of objective fusion evaluation metrics is highlighted to compare the existing methods.

TL;DR: A comprehensive review of techniques incorporating ancillary information in the biometric recognition pipeline is presented in this paper, where the authors provide a comprehensive overview of the role of information fusion in biometrics.

TL;DR: The approach employs a series of learning algorithms including a hybrid approach using Convolutional Neural Network and Long Short-term Memory Recurrent Neural Network on the raw sensor data, eliminating the needs for manual feature extraction and engineering.

TL;DR: A consensus model which considers overconfidence behaviors of LSGDM based on fuzzy preference relations with self-confidence (FPRs-SC) is proposed and a dynamic weight punishment mechanism is implemented for overconfident DMs to improve the consensus efficiently.

TL;DR: The obtained experimental results indicate that the proposed CNNs based network is more accurate and have the better decision map without post-processing algorithms than the other existing state of the art multi-focus fusion methods which used many post- processing algorithms.

TL;DR: This paper reviews the consistency measurements of the different types of RPRs and classified them into four main types: consistency improving methods; consistency-based methods to manage incomplete R PRs; consistency control in consensus decision making methods; and consistency-driven linguistic decision making Methods.

TL;DR: This contribution addresses two main issues by bringing together both decision Making approaches and opinion dynamics to develop a similarity-confidence-consistency based Social network that enables the agents to provide their opinions with the possibility of allocating uncertainty by means of the Intuitionistic fuzzy preference relations and at the same time interact with like-minded agents in order to achieve an agreement.

TL;DR: Zhang et al. as discussed by the authors incorporated illumination information into two-stream deep convolutional neural networks to learn multispectral human-related features under different illumination conditions (daytime and nighttime).

TL;DR: A new score function of HFLTS is proposed to eliminate the defects of the subscript-based operations on HFLtss and shows many advantages over the existing score function in terms of representing both the balanced and unbalanced linguistic information with hesitant degree and linguistic scale functions.

TL;DR: A novel adaptive closed-loop control system and speed up searches model to improve the monitor and control efficiency in IoT networks, specially those which are based in blockchain.

TL;DR: The literature on deriving decision makers’ weights is reviewed to present the state-of-the-art in the group decision making environment and a new classification system is proposed.

TL;DR: A novel cross-modality interactive attention network that takes full advantage of the interactive properties of multispectral input sources is proposed that achieves state-of-the-art performance with high efficiency.

TL;DR: A Social network analysis-based Conflict Relationship Investigation Process (S-CRIP) is presented to detect the conflict relationships among DMs for LSDM events, in which sparse representation is used and three processes constitute the S-CRIP and CD-CRIP-based LSDM model, which is suitable for any numerical representations.

TL;DR: OmniDroid is presented, a large and comprehensive dataset of features extracted from 22,000 real malware and goodware samples aiming to help anti-malware tools creators and researchers when improving, or developing, new mechanisms and tools for Android malware detection.

TL;DR: An adaptive method for gait detection is presented, which models human gait with a hidden Markov model (HMM), and employs a neural network (NN) to deal with the raw measurements and feed the HMM with classifications.