Subjects -> ENGINEERING (Total: 2791 journals)     - CHEMICAL ENGINEERING (248 journals)    - CIVIL ENGINEERING (242 journals)    - ELECTRICAL ENGINEERING (176 journals)    - ENGINEERING (1402 journals)    - ENGINEERING MECHANICS AND MATERIALS (452 journals)    - HYDRAULIC ENGINEERING (56 journals)    - INDUSTRIAL ENGINEERING (100 journals)    - MECHANICAL ENGINEERING (115 journals) ENGINEERING (1402 journals)                  1 2 3 4 5 6 7 8 | Last
 Showing 1 - 200 of 1205 Journals sorted alphabetically 3 Biotech       (Followers: 2) 3D Research       (Followers: 17) AAPG Bulletin       (Followers: 9) Abstract and Applied Analysis       (Followers: 1) Aceh International Journal of Science and Technology       (Followers: 3) ACS Nano       (Followers: 189) Acta Geotechnica       (Followers: 6) Acta Metallurgica Sinica (English Letters)       (Followers: 8) Acta Nova Acta Polytechnica : Journal of Advanced Engineering Acta Universitatis Cibiniensis. Technical Series       (Followers: 1) Active and Passive Electronic Components       (Followers: 5) Additive Manufacturing Letters       (Followers: 7) Adsorption       (Followers: 4) Advanced Energy and Sustainability Research       (Followers: 4) Advanced Engineering Forum       (Followers: 10) Advanced Engineering Research Advanced Journal of Graduate Research       (Followers: 1) Advanced Quantum Technologies       (Followers: 1) Advanced Science       (Followers: 12) Advanced Science Focus       (Followers: 5) Advanced Science Letters       (Followers: 9) Advanced Science, Engineering and Medicine       (Followers: 3) Advanced Synthesis & Catalysis       (Followers: 19) Advanced Theory and Simulations       (Followers: 2) Advances in Applied Energy       (Followers: 6) Advances in Catalysis       (Followers: 7) Advances in Complex Systems       (Followers: 10) Advances in Engineering Software       (Followers: 25) Advances in Fuzzy Systems       (Followers: 5) Advances in Geosciences (ADGEO)       (Followers: 19) Advances in Heat Transfer       (Followers: 27) Advances in Natural Sciences : Nanoscience and Nanotechnology       (Followers: 28) Advances in Operations Research       (Followers: 13) Advances in OptoElectronics       (Followers: 6) Advances in Physics Theories and Applications       (Followers: 12) Advances in Polymer Science       (Followers: 50) Advances in Remote Sensing       (Followers: 58) Advances in Science and Research (ASR)       (Followers: 8) Aerobiologia       (Followers: 2) Aerospace Systems       (Followers: 7) African Journal of Science, Technology, Innovation and Development       (Followers: 7) AIChE Journal       (Followers: 31) Ain Shams Engineering Journal       (Followers: 1) Al-Nahrain Journal for Engineering Sciences Al-Qadisiya Journal for Engineering Sciences AL-Rafdain Engineering Journal Alexandria Engineering Journal       (Followers: 1) AMB Express       (Followers: 1) American Journal of Applied Sciences       (Followers: 21) American Journal of Engineering and Applied Sciences       (Followers: 7) American Journal of Engineering Education       (Followers: 13) American Journal of Environmental Engineering       (Followers: 6) American Journal of Industrial and Business Management       (Followers: 23) Annals of Civil and Environmental Engineering       (Followers: 2) Annals of Combinatorics       (Followers: 3) Annals of Pure and Applied Logic       (Followers: 4) Annals of Regional Science       (Followers: 7) Annals of Science       (Followers: 9) Annual Journal of Technical University of Varna Antarctic Science       (Followers: 1) Applicable Algebra in Engineering, Communication and Computing       (Followers: 3) Applicable Analysis: An International Journal       (Followers: 1) Applications in Energy and Combustion Science       (Followers: 2) Applications in Engineering Science Applied Catalysis A: General       (Followers: 7) Applied Catalysis B: Environmental       (Followers: 9) Applied Clay Science       (Followers: 6) Applied Computational Intelligence and Soft Computing       (Followers: 16) Applied Energy       (Followers: 25) Applied Engineering Letters Applied Magnetic Resonance       (Followers: 3) Applied Nanoscience       (Followers: 7) Applied Network Science       (Followers: 2) Applied Numerical Mathematics       (Followers: 4) Applied Physics Research       (Followers: 5) Applied Spatial Analysis and Policy       (Followers: 5) Arab Journal of Basic and Applied Sciences Arabian Journal for Science and Engineering       (Followers: 1) Archives of Computational Methods in Engineering       (Followers: 5) Archives of Foundry Engineering Archives of Thermodynamics       (Followers: 10) Arctic Arid Zone Journal of Engineering, Technology and Environment ArtefaCToS : Revista de estudios sobre la ciencia y la tecnología Asian Journal of Applied Science and Engineering Asian Journal of Applied Sciences       (Followers: 2) Asian Journal of Biotechnology       (Followers: 8) Asian Journal of Control Asian Journal of Technology Innovation       (Followers: 5) Assembly Automation       (Followers: 2) ATZagenda ATZextra worldwide AURUM : Mühendislik Sistemleri ve Mimarlık Dergisi = Aurum Journal of Engineering Systems and Architecture       (Followers: 1) Australasian Journal of Engineering Education       (Followers: 3) Australasian Physical & Engineering Sciences in Medicine       (Followers: 1) Australian Journal of Multi-Disciplinary Engineering Autocracy : Jurnal Otomasi, Kendali, dan Aplikasi Industri Automotive and Engine Technology Automotive Experiences Automotive Innovation Avances en Ciencias e Ingenierías Avances: Investigación en Ingeniería Balkan Region Conference on Engineering and Business Education       (Followers: 2) Bangladesh Journal of Scientific and Industrial Research Basin Research       (Followers: 6) Batteries       (Followers: 8) Batteries & Supercaps       (Followers: 5) Bautechnik       (Followers: 1) Bell Labs Technical Journal       (Followers: 27) Beni-Suef University Journal of Basic and Applied Sciences Beyond : Undergraduate Research Journal Bhakti Persada : Jurnal Aplikasi IPTEKS Bharatiya Vaigyanik evam Audyogik Anusandhan Patrika (BVAAP) Bilge International Journal of Science and Technology Research       (Followers: 1) Biointerphases       (Followers: 1) Biomaterials Science       (Followers: 11) Biomedical Engineering       (Followers: 11) Biomedical Engineering Letters       (Followers: 3) Biomedical Engineering: Applications, Basis and Communications       (Followers: 4) Biomedical Microdevices       (Followers: 8) Biomedical Science and Engineering       (Followers: 4) Biomicrofluidics       (Followers: 7) Biotechnology Progress       (Followers: 42) Black Sea Journal of Engineering and Science Botswana Journal of Technology       (Followers: 1) Boundary Value Problems Bulletin of Canadian Petroleum Geology       (Followers: 12) Bulletin of Engineering Geology and the Environment       (Followers: 15) Cahiers Droit, Sciences & Technologies       (Followers: 1) Calphad Canadian Geotechnical Journal       (Followers: 28) Canadian Journal of Remote Sensing       (Followers: 51) Carbon Resources Conversion       (Followers: 2) Carpathian Journal of Electronic and Computer Engineering Case Studies in Thermal Engineering       (Followers: 9) Catalysis Communications       (Followers: 7) Catalysis Letters       (Followers: 3) Catalysis Reviews: Science and Engineering       (Followers: 9) Catalysis Science and Technology       (Followers: 9) Catalysis Surveys from Asia       (Followers: 4) Catalysis Today       (Followers: 4) CEAS Space Journal       (Followers: 6) Cell Reports Physical Science Cellular and Molecular Neurobiology       (Followers: 2) CFD Letters       (Followers: 7) Chaos : An Interdisciplinary Journal of Nonlinear Science       (Followers: 3) Chaos, Solitons & Fractals       (Followers: 1) Chaos, Solitons & Fractals : X       (Followers: 1) Chinese Journal of Catalysis       (Followers: 2) Chinese Journal of Engineering       (Followers: 1) Chinese Journal of Population, Resources and Environment Chinese Science Bulletin Ciencia e Ingenieria Neogranadina Ciencia en su PC       (Followers: 1) Ciencia y Tecnología Ciencias Holguin       (Followers: 1) CienciaUAT Cientifica CIRP Annals - Manufacturing Technology       (Followers: 10) CIRP Journal of Manufacturing Science and Technology       (Followers: 12) City, Culture and Society       (Followers: 23) Clay Minerals       (Followers: 7) Cleaner Engineering and Technology       (Followers: 5) Cleaner Environmental Systems       (Followers: 5) Coastal Engineering       (Followers: 16) Coastal Engineering Journal       (Followers: 7) Coastal Engineering Proceedings : Proceedings of the International Conference on Coastal Engineering       (Followers: 1) Coastal Management       (Followers: 29) Coatings       (Followers: 2) Cogent Engineering       (Followers: 1) Cognitive Computation       (Followers: 2) Color Research & Application       (Followers: 1) COMBINATORICA Combustion Theory and Modelling       (Followers: 18) Combustion, Explosion, and Shock Waves       (Followers: 21) Communications Faculty of Sciences University of Ankara Series A2-A3 Physical Sciences and Engineering Communications in Numerical Methods in Engineering       (Followers: 2) Components, Packaging and Manufacturing Technology, IEEE Transactions on       (Followers: 27) Composite Interfaces       (Followers: 6) Composite Structures       (Followers: 246) Composites Part A : Applied Science and Manufacturing       (Followers: 179) Composites Part B : Engineering       (Followers: 224) Composites Part C : Open Access       (Followers: 1) Composites Science and Technology       (Followers: 152) Comptes Rendus : Mécanique       (Followers: 2) Computation       (Followers: 1) Computational Geosciences       (Followers: 17) Computational Optimization and Applications       (Followers: 9) Computer Applications in Engineering Education       (Followers: 6) Computer Science and Engineering       (Followers: 15) Computers & Geosciences       (Followers: 30) Computers & Mathematics with Applications       (Followers: 8) Computers and Electronics in Agriculture       (Followers: 7) Computers and Geotechnics       (Followers: 11) Computing and Visualization in Science       (Followers: 6) Computing in Science & Engineering       (Followers: 31) Conciencia Tecnologica Continuum Mechanics and Thermodynamics       (Followers: 8) Control Engineering Practice       (Followers: 46)
Similar Journals
 Cognitive ComputationJournal Prestige (SJR): 0.908 Citation Impact (citeScore): 4Number of Followers: 2      Hybrid journal (It can contain Open Access articles) ISSN (Print) 1866-9964 - ISSN (Online) 1866-9956 Published by Springer-Verlag  [2469 journals]
• A Calibration-free Approach to Implementing P300-based
Brain–computer Interface

• Free pre-print version: Loading...

Abstract: Introduction: As a direct bridge between the brain and the outer world, brain–computer interface (BCI) is expected to replace, restore, enhance, supplement, or improve the natural output of brain. The prospect of BCI serving humans is very broad. However, the extensive applications of BCI have not been fully achieved. One of reasons is that the cost of calibration reduces the convenience and usability of BCI. Methods: In this study, we proposed a calibration-free approach, which is based on the ideas of reinforcement learning and transfer learning, for P300-based BCI. This approach, composed of two algorithms: P300 linear upper confidence bound (PLUCB) and transferred PLUCB (TPLUCB), is able to learn during the usage by exploration and exploitation and allows P300-based BCI to start working without any calibration. Results: We tested the performances of PLUCB and TPLUCB using stepwise linear discriminant analysis (SWLDA), a commonly used method that needs calibration, as a baseline in simulated online experiments. The results showed the merits of PLUCB and TPLUCB. PLUCB can quickly increase the accuracies to the level of SWLDA. TPLUCB has surpassed SWLDA in the sample accuracy since it starts running. Both PLUCB and TPLUCB have the ability to keep improving the classification performance during the process. The overall sample accuracies ( $$73.6\pm 4.8\%$$ , $$73.1\pm 4.9\%$$ ), overall symbol accuracies ( $$80.4\pm 12.8\%$$ , $$79.6\pm 14.0\%$$ ), F-measures ( $$0.45\pm 0.06$$ , $$0.44\pm 0.06$$ ) and information transfer ratios (ITR) ( $$36.4\pm 9.1$$ , $$35.5\pm 9.8$$ ) of PLUCB and TPLUCB are significantly better than those of SWLDA (overall sample accuracy: $$58.8\pm 3.8\%$$ , overall symbol accuracy: $$69.0\pm 18.3\%$$ , F-measure: $$0.38\pm 0.04$$ , ITR: $$28.7\pm 10.7$$ ). Conclusions: The proposed approach, which does not need calibration but outperform SWLDA, is a very good option for the implementation of P300-based BCI.
PubDate: 2022-01-24

• Centroid Coordinate Ranking of Pythagorean Fuzzy Numbers and its
Application in Group Decision Making

• Free pre-print version: Loading...

Abstract: Cognitive computing contains different cognitive characteristics, especially when dealing with group decision-making problems, it is considered as a cognitive-based human behavior, in which collecting and processing data from multiple resources is an important stage. Intuitionistic fuzzy number (IFN) and Pythagorean fuzzy number (PFN) are the most reliable tools to deal with fuzzy information by utilizing membership and non-membership, where the distance measure and similarity of IFNs or PFNs play an important role in dealing with incomplete information in order to achieve the final decision and PFN is a generalization of IFN. Motivated by these, some important concepts for PFNs are proposed by geometric methods to deal with group decision-making problems in this paper. Through counter examples, it is pointed out that the score function and accuracy function of PFNs are inconsistent with the traditional ranking rules of IFNs, the concepts of the centroid coordinate and hesitation factor are proposed by geometric distance. In addition, Pythagorean fuzzy distance measure (PFDM) through the centroid coordinate and hesitation factor are introduced, and proved the distance measure satisfies the axiomatic conditions of distance, a calculation example is given in the form of tables. A unified ranking method for IFNs and PFNs is given by comparing with the smallest PFN (0,1). The weight vector, positive or negative ideal solution is calculated by aggregating centroid coordinate matrices, a new TOPSIS method is given by using Pythagorean fuzzy weighted distance (PFWD) and relative closeness. These results show that the decision matrix and positive (negative) ideal solutions represented by the centroid coordinate and hesitation factor can reflect the fuzzy information more comprehensively. The proposed method not only has a wide range of application, but also reduces the loss of information and is easier to be implemented. It is only applied to the multi criteria decision making problem for the first time, it also has some other good properties that need to be further explored and supplemented. This provides a theoretical basis for studying the wide application of Pythagorean fuzzy sets.
PubDate: 2022-01-24

• Latent Space Exploration and Functionalization of a Gated Working Memory
Model Using Conceptors

• Free pre-print version: Loading...

Abstract: Working memory is the ability to maintain and manipulate information. We introduce a method based on conceptors that allows us to manipulate information stored in the dynamics (latent space) of a gated working memory model. This latter model is based on a reservoir: a random recurrent network with trainable readouts. It is trained to hold a value in memory given an input stream when a gate signal is on and to maintain this information when the gate is off. The memorized information results in complex dynamics inside the reservoir that can be faithfully captured by a conceptor. Such conceptors allow us to explicitly manipulate this information in order to perform various, but not arbitrary, operations. In this work, we show (1) how working memory can be stabilized or discretized using such conceptors, (2) how such conceptors can be linearly combined to form new memories, and (3) how these conceptors can be extended to a functional role. These preliminary results suggest that conceptors can be used to manipulate the latent space of the working memory even though several results we introduce are not as intuitive as one would expect.
PubDate: 2022-01-22

• A Bibliometric Study and Science Mapping Research of Intelligent Decision

• Free pre-print version: Loading...

Abstract: Intelligent decision (ID) has received a great deal of attention and has been integrated into various fields, such as machine learning, fuzzy inference system, and natural language processing. The advanced technologies have become hot topics and have been made great development and innovations in academic documents. This paper is a comprehensive review in the field of ID based on bibliometric analysis and strategic analysis. First, the descriptive statistics and results are presented, including database, annual publications, research directions, and hotspots. Based on the visualization tools (including VOS viewer, CiteSpace, Bibexcel, and GPS visualizer), from the perspective of the author keyword, the current research topics, and the development evolution are presented. Some bibliometric analysis methods are applied, such as co-occurrence analysis, timeline view analysis, and burst detection analysis. Then, this paper identifies the most influential countries/regions, institutions, and authors. Next, some important themes are further discussed by strategic analysis and overlapping analysis. This paper helps scholars with understanding the development trajectory and statistical model of ID research to promote in-depth exploration.
PubDate: 2022-01-22

• FFNet: Feature Fusion Network for Few-shot Semantic Segmentation

• Free pre-print version: Loading...

Abstract: Semantic segmentation aims at assigning a category label to each pixel in an image. Deep neural networks have achieved many breakthrough research achievements on this task. Nevertheless, there exist two critical bottleneck problems to be solved. First, deep neural networks usually need to be trained on large-scale labeled datasets, which are expensive to obtain or label. Second, traditional semantic segmentation methods are difficult to predict unseen classes after training. To address these problems, few-shot semantic segmentation is proposed, and recent methods have achieved impressive performance. However, many of the existing approaches ignore the semantic correlation between data and fail to generate discriminative features for the semantic segmentation. In this paper, to address the above issue, we propose a feature fusion network (FFNet) for few-shot semantic segmentation to enhance the discriminative ability of the learned data representations. Specifically, a task attention module is devised to learn the semantic correlation between data. Then, a multi-scale feature fusion module is trained to adaptively fuse the contextual information at multiple scale, thus capturing multi-scale object information. To the end, the proposed FFNet experiments conducted on the PASCAL- $$5^i$$ and COCO- $$20^i$$ datasets demonstrate the superiority of our proposed FFNet and show its advantage over existing approaches.
PubDate: 2022-01-22

• HAKE: an Unsupervised Approach to Automatic Keyphrase Extraction for
Multiple Domains

• Free pre-print version: Loading...

Abstract: Keyphrases capture the main content of a free text document. The task of automatic keyphrase extraction (AKPE) plays a significant role in retrieving and summarizing valuable information from several documents with different domains. Various techniques have been proposed for this task. However, supervised AKPE requires large annotated data and depends on the tested domain. An alternative solution is to consider a new independent domain method that can be applied to several domains (such as medical, social). In this paper, we tackle keyphrase extraction from single documents with HAKE, a novel unsupervised method that takes full advantage of mining linguistic, statistical, structural, and semantic text features simultaneously to select the most relevant keyphrases in a text. HAKE achieves higher F-scores than the unsupervised state-of-the-art systems on standard datasets and is suitable for real-time processing of large amounts of Web and text data across different domains. With HAKE, we also explicitly increase coverage and diversity among the selected keyphrases by introducing a novel technique (based on a parse tree approach, part of speech tagging, and filtering) for candidate keyphrase identification and extraction. This technique allows us to generate a comprehensive and meaningful list of candidate keyphrases and reduce the candidate set’s size without increasing the computational complexity. HAKE’s effectiveness is compared to twelve state-of-the-art and recent unsupervised approaches, as well as to some other supervised approaches. Experimental analysis is conducted to validate the proposed method using five of the top available benchmark corpora from different domains and shows that HAKE significantly outperforms both the existing unsupervised and supervised methods. Our method does not require training on a particular set of documents, nor does it depend on external corpora, dictionaries, domain, or text size. Our experiments confirm that HAKE’s candidate selection model and its ranking model are effective.
PubDate: 2022-01-21

• Adaptive Three-Way C-Means Clustering Based on the Cognition of Distance
Stability

• Free pre-print version: Loading...

Abstract: Soft clustering can be regarded as a cognitive computing method that seeks to deal with the clustering with fuzzy boundary. As a classical soft clustering algorithm, rough k-means (RKM) has yielded various extensions. However, some challenges remain in existing RKM extensions. On the one hand, the user-defined cutoff threshold is subjective and cannot be changed during iteration. On the other hand, the weight of the object to the cluster center is calculated by membership grade and a subjective parameter, that is, the fuzzifier, which complicates the issue and reduces the robustness of the algorithm. In this paper, inspired by human cognition of distance stability, an adaptive three-way c-means algorithm is proposed. First, in human cognition, objects are clustered according to the stability of their distance to the clusters, and variance is an effective way to measure the stability of data. Based on this, an adaptive cutoff threshold is introduced by determining the maximum increment between the variances of distance. Second, based on the cognition that distance is inversely proportional to weight, the weight equation is defined by distance without introducing any subjective parameters. Then, combined with the adaptive cutoff threshold and weight equation, A-3WCM is proposed. The experimental results show that A-3WCM exhibits excellent performance and outperforms five representative algorithms related to RKM on nine popular datasets.
PubDate: 2022-01-21

• A Novel Multiple Feature-Based Engine Knock Detection System using Sparse
Bayesian Extreme Learning Machine

• Free pre-print version: Loading...

Abstract: Automotive engine knock is an abnormal combustion phenomenon that affects engine performance and lifetime expectancy, but it is difficult to detect. Collecting engine vibration signals from an engine cylinder block is an effective way to detect engine knock. This paper proposes an intelligent engine knock detection system based on engine vibration signals. First, filtered signals are obtained by utilizing variational mode decomposition (VMD), which decomposes the original time domain signals into a series of intrinsic mode functions (IMFs). Moreover, the values of the balancing parameter and the number of IMF modes are optimized using genetic algorithm (GA). IMFs with sample entropy higher than the mean are then selected as sensitive subcomponents for signal reconstruction and subsequently removed. A multiple feature learning approach that considers time domain statistical analysis (TDSA), multi-fractal detrended fluctuation analysis (MFDFA) and alpha stable distribution (ASD) simultaneously, is utilized to extract features from the denoised signals. Finally, the extracted features are trained by sparse Bayesian extreme learning machine (SBELM) to overcome the sensitivity of hyperparameters in conventional machine learning algorithms. A test rig is designed to collect the raw engine data. Compared with other technology combinations, the optimal scheme from signal processing to feature classification is obtained, and the classification accuracy of the proposed integrated engine knock detection method can achieve 98.27%. We successfully propose and test a universal intelligence solution for the detection task.
PubDate: 2022-01-19

• Observer and Command-Filter-Based Adaptive Neural Network Control
Algorithms for Nonlinear Multi-agent Systems with Input Delay

• Free pre-print version: Loading...

Abstract: Over the last decades, many researchers have investigated the distributed adaptive consensus tacking control algorithm of multi-agent systems (MASs). Nevertheless, the existing works involving the command-filter-based adaptive consensus problem for nonlinear multi-agent systems subjected to the unmeasurable states are relatively few. Besides that, the immeasurable states and the input delay will bring few challenging in dealing with the consensus problem for MASs. (1) The radial basis function neural networks (RBF NNs) are utilized to approximate the unknown nonlinear functions and the NN-based observer is established to copy with the unmeasurable states. (2) The backstepping design method of distributed adaptive consensus control is put forward on basis of the command filtering method, which overcomes the complexity explosion problem and eliminates errors by introducing compensation signals. (3) The Pade approximation approach is served to remove the obstacle originating from the input delay. This paper addresses the observer and command-filter-based adaptive tracking control problem for nonlinear multi-agent systems with the unmeasurable states and input delay under the directed graph. The Lyapunov stability theory is utilized to prove that the proposed approach can ensure that all signals in the closed-loop system reach cooperatively semi-globally uniformly ultimately bounded (CSUUB). The simulation result is presented, and it further manifests that the effectiveness of this scheme.
PubDate: 2022-01-18

• Stability Analysis of Stochastic Delayed Differential Systems with
State-Dependent-Delay Impulses: Application of Neural Networks

• Free pre-print version: Loading...

Abstract: The previously studied stochastic delayed nonlinear systems with constant delayed or time-varying delayed impulsive controller, while we study stochastic delayed nonlinear systems under state-dependent-delayed impulsive controller. The difficulty is how to determine the time of impulses occurrence. Employing the Halanay differential inequality, Itô’s formula, the average impulsive interval, impulsive control theory, comparison properties, several effective conditions ensuring stability of stochastic delayed nonlinear systems under state-dependent delayed impulsive controller are derived. This paper contributed to the stability analysis of delayed impulsive nonlinear systems with stochastic perturbation, which the impulsive involved delay is state dependent. We have developed exponential stability of delayed nonlinear systems with state-dependent-delay impulses and stochastic disturbance in this paper. In the future, more new methods should also be proposed. At the same time, we will consider the nonlinear systems with unbounded delays.
PubDate: 2022-01-17

• PS-Net: Progressive Selection Network for Salient Object Detection

• Free pre-print version: Loading...

Abstract: Low-level features contain abundant details and high-level features have rich semantic information. Integrating multi-scale features in an appropriate way is significant for salient object detection. However, direct concatenation or addition taken by most methods ignores the distinctions of contribution among multi-scale features. Besides, most salient object detection models fail to dynamically adjust receptive fields to fit objects of various sizes. To tackle these problems, we propose a Progressive Selection Network (PS-Net). Specifically, PS-Net dynamically extracts high-level features and encourages high-level features to guide low-level features to suppress the background response of the original features. We proposed a salient model PS-Net that selects features progressively at multiply levels. First, we propose a Pyramid Feature Dynamic Extraction module to dynamically select appropriate receptive fields to extract high-level features by Feature Dynamic Extraction modules step by step. Besides, a Self-Interaction Attention module is designed to extract detailed information for low-level features. Finally, we design a Scale Aware Fusion module to fuse these multiple features for adequate exploitation of high-level features to refine low-level features gradually. Compared with 19 start-of-the-art methods on 6 public benchmark datasets, the proposed method achieves remarkable performance in both quantitative and qualitative evaluation. We performed a lot of ablation studies, and more discussions to demonstrate the effectiveness and superiority of our proposed method. In this paper, we propose a PS-Net for effective salient object detection. Extensive experiments on 6 datasets validate that the proposed model outperforms 19 state-of-the-art methods under different evaluation metrics.
PubDate: 2022-01-16

• Road Segmentation from High-Fidelity Remote Sensing Images using a Context
Information Capture Network

• Free pre-print version: Loading...

Abstract: The automatic extraction of roads or buildings from remote sensing imagery plays a significant role in many urban applications. Recently, due to the impressive performance of deep learning, various road segmentation methods based on the fully convolutional network (FCN) have been proposed for optical remote sensing images. However, the existing FCN-based high-fidelity remote sensing image segmentation methods still have some limitations. As the repeated convolution and pooling operations employed in an FCN reduce the feature resolution and lose some detailed information, FCNs have a limited capacity to mine long-range dependencies among features. To address this issue, a context information capture network (CM-FCN) for road segmentation is proposed. To capture and aggregate multiscale contextual information, a dilated convolution module is designed. Furthermore, to boost the long-range dependencies of features for road detection, two attention modules employing the attention mechanism to adaptively combine local features with their global dependencies are designed. The context features extracted from the dilated convolution module are then fused into the attention modules to further improve the segmentation performance. The proposed model is evaluated on three challenging remote sensing image road segmentation datasets and one building segmentation dataset, including a dataset with our own manual labels. Comparisons demonstrate the effectiveness of our proposed method. We conclude that our proposed CM-FCN has the potential to automatically segment roads and buildings from high-resolution remote sensing images with an accuracy that renders it a useful tool for practical application scenarios.
PubDate: 2022-01-15

• LightAdam: Towards a Fast and Accurate Adaptive Momentum Online Algorithm

• Free pre-print version: Loading...

Abstract: Adaptive optimization algorithms enjoy fast convergence and have been widely exploited in pattern recognition and cognitively-inspired machine learning. These algorithms may however be of high computational cost and low generalization ability due to their projection steps. Such limitations make them difficult to be applied in big data analytics, which may typically be seen in cognitively inspired learning, e.g. deep learning tasks. In this paper, we propose a fast and accurate adaptive momentum online algorithm, called LightAdam, to alleviate the drawbacks of projection steps for the adaptive algorithms. The proposed algorithm substantially reduces computational cost for each iteration step by replacing high-order projection operators with one-dimensional linear searches. Moreover, we introduce a novel second-order momentum and engage dynamic learning rate bounds in the proposed algorithm, thereby obtaining a higher generalization ability than other adaptive algorithms. We theoretically analyze that our proposed algorithm has a guaranteed convergence bound, and prove that our proposed algorithm has better generalization capability as compared to Adam. We conduct extensive experiments on three public datasets for image pattern classification, and validate the computational benefit and accuracy performance of the proposed algorithm in comparison with other state-of-the-art adaptive optimization algorithms
PubDate: 2022-01-11

• Deep Learning for Reliable Classification of COVID-19, MERS, and SARS from
Chest X-ray Images

• Free pre-print version: Loading...

Abstract: Novel coronavirus disease (COVID-19) is an extremely contagious and quickly spreading coronavirus infestation. Severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS), which outbreak in 2002 and 2011, and the current COVID-19 pandemic are all from the same family of coronavirus. This work aims to classify COVID-19, SARS, and MERS chest X-ray (CXR) images using deep convolutional neural networks (CNNs). To the best of our knowledge, this classification scheme has never been investigated in the literature. A unique database was created, so-called QU-COVID-family, consisting of 423 COVID-19, 144 MERS, and 134 SARS CXR images. Besides, a robust COVID-19 recognition system was proposed to identify lung regions using a CNN segmentation model (U-Net), and then classify the segmented lung images as COVID-19, MERS, or SARS using a pre-trained CNN classifier. Furthermore, the Score-CAM visualization method was utilized to visualize classification output and understand the reasoning behind the decision of deep CNNs. Several deep learning classifiers were trained and tested; four outperforming algorithms were reported: SqueezeNet, ResNet18, InceptionV3, and DenseNet201. Original and preprocessed images were used individually and all together as the input(s) to the networks. Two recognition schemes were considered: plain CXR classification and segmented CXR classification. For plain CXRs, it was observed that InceptionV3 outperforms other networks with a 3-channel scheme and achieves sensitivities of 99.5%, 93.1%, and 97% for classifying COVID-19, MERS, and SARS images, respectively. In contrast, for segmented CXRs, InceptionV3 outperformed using the original CXR dataset and achieved sensitivities of 96.94%, 79.68%, and 90.26% for classifying COVID-19, MERS, and SARS images, respectively. The classification performance degrades with segmented CXRs compared to plain CXRs. However, the results are more reliable as the network learns from the main region of interest, avoiding irrelevant non-lung areas (heart, bones, or text), which was confirmed by the Score-CAM visualization. All networks showed high COVID-19 detection sensitivity (> 96%) with the segmented lung images. This indicates the unique radiographic signature of COVID-19 cases in the eyes of AI, which is often a challenging task for medical doctors.
PubDate: 2022-01-11

• A Fuzzy Collaborative Intelligence Approach to Group Decision-Making: a
Case Study of Post-COVID-19 Restaurant Transformation

• Free pre-print version: Loading...

Abstract: In a fuzzy group decision-making task, when decision makers lack consensus, existing methods either ignore this fact or force a decision maker to modify his/her judgment. However, these actions may be unreasonable. In this study, a fuzzy collaborative intelligence approach that seeks the consensus among experts in a novel way is proposed. Fuzzy collaborative intelligence is the application of biologically inspired fuzzy logic to a group task. The proposed methodology is based on the fact that a decision maker must make a choice even if he/she is uncertain. As a result, the decision maker’s fuzzy judgment matrix may not be able to represent his/her judgment. To solve such a problem, the fuzzy judgment matrix of each decision maker is decomposed into several fuzzy judgment submatrices. From the fuzzy judgment submatrices of all decision makers, a consensus can be easily identified. The proposed fuzzy collaborative intelligence approach and several existing methods have been applied to the case of the post-COVID-19 transformation of a Japanese restaurant in Taiwan. Because such transformation was beyond the expectation of the Japanese restaurant, the employees lacked consensus if existing methods were applied to identify their consensus. The proposed methodology solved this problem. The optimal transformation plan involved increasing the distance between tables, erecting screens between tables, and improving air circulation. In a fuzzy group decision-making task, an acceptable decision cannot be made without the consensus among decision makers. Ignoring this or forcing decision makers to modify their preferences is unreasonable. Identifying the consensus among experts from another point of view is a viable treatment.
PubDate: 2022-01-10

• Automatic Assessment of Motor Impairments in Autism Spectrum Disorders: A
Systematic Review

• Free pre-print version: Loading...

Abstract: Autism spectrum disorder (ASD) is mainly described as a disorder of communication and socialization. However, motor abnormalities are also common in ASD. New technologies may offer quantitative and automatic metrics to measure movement difficulties. We sought to identify computational methods to automatize the assessment of motor impairments in ASD. We systematically searched for the terms ’autism’, ’movement’, ’automatic’, ’computational’ and ’engineering’ in IEEE (Institute of Electrical and Electronics Engineers), Medline and Scopus databases and reviewed the literature from inception to 2018. We included all articles discussing: (1) automatic assessment/new technologies, (2) motor behaviours and (3) children with ASD. We excluded studies that included patient’s or parent’s reported outcomes as online questionnaires that focused on computational models of movement, but also eye tracking, facial emotion or sleep. In total, we found 53 relevant articles that explored static and kinetic equilibrium, like posture, walking, fine motor skills, motor synchrony and movements during social interaction that can be impaired in individuals with autism. Several devices were used to capture relevant motor information such as cameras, 3D cameras, motion capture systems, accelerometers. Interestingly, since 2012, the number of studies increased dramatically as technologies became less invasive, more precise and more affordable. Open-source software has enabled the extraction of relevant data. In a few cases, these technologies have been implemented in serious games, like “Pictogram Room”, to measure the motor status and the progress of children with ASD. Movement computing opens new perspectives for patient assessment in ASD research, enabling precise characterizations in experimental and at-home settings, and a better understanding of the role of sensorimotor disturbances in the development of social cognition and ASD. These methods would likely enable researchers and clinicians to better distinguish ASD from other motors disorders while facilitating an improved monitoring of children’s progress in more ecological settings (i.e. at home or school).
PubDate: 2022-01-10

• An Attention-Driven Multi-label Image Classification with Semantic
Embedding and Graph Convolutional Networks

• Free pre-print version: Loading...

Abstract: Multi-label image classification is a fundamental and vital task in computer vision. The latest methods are mostly based on deep learning and exhibit excellent performance in understanding images. However, in previous studies, only capture the image content information has been captured using convolutional neural networks (CNNs), and the semantic structure information and implicit dependencies between labels and image regions have been ignored. Therefore, it is necessary to develop more effective methods for integrating semantic information and visual features in multi-label image classification. In this study, we propose a novel framework for multi-label image classification, named FLNet, which simultaneously takes advantage of the visual features and semantic structure. Specifically, to enhance the association between semantic annotations and image regions, we first integrate the attention mechanism with a CNN to focus on the target regions while ignoring other useless surrounding information and then employ graph convolutional network (GCN) to capture the structure information between multiple labels. Based on our architecture, we also introduce the lateral connections to repeatedly inject the label system into the CNN backbone during the GCN learning process to improve performance and, consequently, learn interdependent classifiers for each image label. We apply our method to multi-label image classification. The experiments on two public multi-label benchmark datasets, namely, MS-COCO and PASCAL visual object classes challenge (VOC 2007), demonstrate that our approach outperforms other existing state-of-the-art methods. Our method learns specific target regions and enhances the association between labels and image regions by using semantic information and attention mechanism. Thus, we combine the advantages of both visual and semantic information to further improve the image classification performance. Finally, the correctness and effectiveness of the proposed method are proven by visualizing the classifier results.
PubDate: 2022-01-09

• Dense Tissue Pattern Characterization Using Deep Neural Network

• Free pre-print version: Loading...

Abstract: Breast tumors are from the common infections among women around the world. Classifying the various types of breast tumors contribute to treating breast tumors more efficiently. However, this classification task is often hindered by dense tissue patterns captured in mammograms. The present study has been proposed a dense tissue pattern characterization framework using deep neural network. A total of 322 mammograms belonging to the mini-MIAS dataset and 4880 mammograms from DDSM dataset have been taken, and an ROI of fixed size 224 × 224 pixels from each mammogram has been extracted. In this work, tedious experimentation has been executed using different combinations of training and testing sets using different activation function with AlexNet, ResNet-18 model. Data augmentation has been used to create a similar type of virtual image for proper training of the DL model. After that, the testing set is applied on the trained model to validate the proposed model. During experiments, four different activation functions ‘sigmoid’, ‘tanh’, ‘ReLu’, and ‘leakyReLu’ are used, and the outcome for each function has been reported. It has been found that activation function ‘ReLu’ perform always outstanding with respect to others. For each experiment, classification accuracy and kappa coefficient have been computed. The obtained accuracy and kappa value for MIAS dataset using ResNet-18 model is 91.3% and 0.803, respectively. For DDSM dataset, the accuracy of 92.3% and kappa coefficient value of 0.846 are achieved. After the combination of both dataset images, the achieved accuracy is 91.9%, and kappa coefficient value is 0.839 using ResNet-18 model. Finally, it has been concluded that the ResNet-18 model and ReLu activation function yield outstanding performance for the task.
PubDate: 2022-01-08

• Embeddings Evaluation Using a Novel Measure of Semantic Similarity

• Free pre-print version: Loading...

Abstract: Lexical taxonomies and distributional representations are largely used to support a wide range of NLP applications, including semantic similarity measurements. Recently, several scholars have proposed new approaches to combine those resources into unified representation preserving distributional and knowledge-based lexical features. In this paper, we propose and implement TaxoVec, a novel approach to selecting word embeddings based on their ability to preserve taxonomic similarity. In TaxoVec, we first compute the pairwise semantic similarity between taxonomic words through a new measure we previously developed, the Hierarchical Semantic Similarity (HSS), which we show outperforms previous measures on several benchmark tasks. Then, we train several embedding models on a text corpus and select the best model, that is, the model that maximizes the correlation between the HSS and the cosine similarity of the pair of words that are in both the taxonomy and the corpus. To evaluate TaxoVec, we repeat the embedding selection process using three other semantic similarity benchmark measures. We use the vectors of the four selected embeddings as machine learning model features to perform several NLP tasks. The performances of those tasks constitute an extrinsic evaluation of the criteria for the selection of the best embedding (i.e. the adopted semantic similarity measure). Experimental results show that (i) HSS outperforms state-of-the-art measures for measuring semantic similarity in taxonomy on a benchmark intrinsic evaluation and (ii) the embedding selected through TaxoVec achieves a clear victory against embeddings selected by the competing measures on benchmark NLP tasks. We implemented the HSS, together with other benchmark measures of semantic similarity, as a full-fledged Python package called TaxoSS, whose documentation is available at https://pypi.org/project/TaxoSS.
PubDate: 2022-01-08

• Bifurcation Study for Fractional-Order Three-Layer Neural Networks
Involving Four Time Delays

• Free pre-print version: Loading...

Abstract: During the past several decades, many scholars deal with the stability behavior and Hopf bifurcation phenomenon of fractional-order delayed neural networks. However, the literature involving the stability issue and Hopf bifurcation behavior of fractional-order neural networks with multiple time delays is relatively scarce. This article is principally concerned with the stability problem and of Hopf bifurcation behavior of fractional-order three-layer neural networks involving multiple time delays. Variable substitution, Laplace transform, bifurcation principle of fractional-order dynamical system and computer simulation skill are employed. The delay-independent stability criterion and the sufficient condition of onset of Hopf bifurcation of three-layer neural networks are set up. It shows that if the sum of both different delays passes a key value, then the system loses its stability and the Hopf bifurcation phenomenon will take place. The study manifests that delay plays a most momentous part in stabilizing system and controlling bifurcation behavior for the fractional-order delayed three-layer neural networks. The researchful results of this article are an important theoretical cornerstone in controlling and adjusting neural networks. The obtained conclusions are completely novel and complement the earlier research results.
PubDate: 2022-01-07

JournalTOCs
School of Mathematical and Computer Sciences
Heriot-Watt University
Edinburgh, EH14 4AS, UK
Email: journaltocs@hw.ac.uk
Tel: +00 44 (0)131 4513762

Your IP address: 54.224.117.125

Home (Search)
API
About JournalTOCs
News (blog, publications)

JournalTOCs © 2009-