Yan Zhou, Lili Pan^*, Rongyu Chen, Weizhi Shao

Journal of Information Hiding and Privacy Protection, Vol.2, No.2, pp. 77-86, 2020, DOI:10.32604/jihpp.2020.010486 - 11 November 2020

Abstract In large-scale image retrieval, deep features extracted by Convolutional Neural Network (CNN) can effectively express more image information than those extracted by traditional manual methods. However, the deep feature dimensions obtained by Deep Convolutional Neural Network (DCNN) are too high and redundant, which leads to low retrieval efficiency. We propose a novel image retrieval method, which combines deep features selection with improved DCNN and hash transform based on high-dimension features reduction to gain lowdimension deep features and realizes efficient image retrieval. Firstly, the improved network is based on the existing deep model to build a… More >

Rongyu Chen, Lili Pan^*, Yan Zhou, Qianhui Lei

Journal of Information Hiding and Privacy Protection, Vol.2, No.2, pp. 67-76, 2020, DOI:10.32604/jihpp.2020.010472 - 11 November 2020

Abstract With the rapid development of information technology, the speed and efficiency of image retrieval are increasingly required in many fields, and a compelling image retrieval method is critical for the development of information. Feature extraction based on deep learning has become dominant in image retrieval due to their discrimination more complete, information more complementary and higher precision. However, the high-dimension deep features extracted by CNNs (convolutional neural networks) limits the retrieval efficiency and makes it difficult to satisfy the requirements of existing image retrieval. To solving this problem, the high-dimension feature reduction technology is… More >

Jingfa Li¹, Bo Yu^1,*, Qianqian Shao², Dongliang Sun¹

CMES-Computer Modeling in Engineering & Sciences, Vol.125, No.2, pp. 541-563, 2020, DOI:10.32604/cmes.2020.011835 - 12 October 2020

Abstract In this work, the drag-reducing mechanism of high-Reynoldsnumber turbulent channel flow with surfactant additives is investigated by using large eddy simulation (LES) method. An N-parallel finitely extensible nonlinear elastic model with Peterlin’s approximation (FENE-P) is used to describe the rheological behaviors of non-Newtonian fluid with surfactant. To close the filtered LES equations, a hybrid subgrid scale (SGS) model coupling the spatial filter and temporal filter is applied to compute the subgrid stress and other subfilter terms. The finite difference method and projection algorithm are adopted to solve the LES governing equations. To validate the correctness More >

Soukaina Mjahed^1,*, Khadija Bouzaachane¹, Ahmad Taher Azar^2,3, Salah El Hadaj¹, Said Raghay¹

CMES-Computer Modeling in Engineering & Sciences, Vol.125, No.2, pp. 459-494, 2020, DOI:10.32604/cmes.2020.010791 - 12 October 2020

Abstract This paper focuses on the unsupervised detection of the Higgs boson particle using the most informative features and variables which characterize the “Higgs machine learning challenge 2014” data set. This unsupervised detection goes in this paper analysis through 4 steps: (1) selection of the most informative features from the considered data; (2) definition of the number of clusters based on the elbow criterion. The experimental results showed that the optimal number of clusters that group the considered data in an unsupervised manner corresponds to 2 clusters; (3) proposition of a new approach for hybridization of… More >

Sun-Taag Choe, We-Duke Cho^*, Jai-Hoon Kim, and Ki-Hyung Kim

Intelligent Automation & Soft Computing, Vol.26, No.4, pp. 679-691, 2020, DOI:10.32604/iasc.2020.010102

Abstract Recent research on activity recognition in wearable devices has identified a key challenge: k-nearest neighbors (k-NN) algorithms have a high operational time complexity. Thus, these algorithms are difficult to utilize in embedded wearable devices. Herein, we propose a method for reducing this complexity. We apply a clustering algorithm for learning data and assign labels to each cluster according to the maximum likelihood. Experimental results show that the proposed method achieves effective operational levels for implementation in embedded devices; however, the accuracy is slightly lower than that of a traditional k-NN algorithm. Additionally, our method provides More >

Yanqiu Cao¹, Chao Tan¹, Genlin Ji^{1, *}

Journal on Big Data, Vol.2, No.1, pp. 19-31, 2020, DOI:10.32604/jbd.2020.01003 - 07 September 2020

Abstract In the last few years, smartphone usage and driver sleepiness have been unanimously considered to lead to numerous road accidents, which causes many scholars to pay attention to autonomous driving. For this complexity scene, one of the major challenges is mining information comprehensively from massive features in vehicle video. This paper proposes a multi-label classification method MCM-VV (Multi-label Classification Method for Vehicle Video) for vehicle video to judge the label of road condition for unmanned system. Method MCM-VV includes a process of feature extraction and a process of multi-label classification. During feature extraction, grayscale, lane… More >

Jeong-Sik Park¹, Seok-Hoon Kim^2,*

Intelligent Automation & Soft Computing, Vol.26, No.1, pp. 149-159, 2020, DOI:10.31209/2019.100000136

Abstract Variety types of smart home devices have a main function of a human-machine interaction by speech recognition. Speech recognition system may be vulnerable to rapidly changing noises in home environments. This study proposes an efficient noise cancellation approach to eliminate the noises directly on the devices in real time. Firstly, we propose an advanced voice activity detection (VAD) technique to efficiently detect speech and non-speech regions on the basis of spectral property of speech signals. The VAD is then employed to enhance the conventional spectral subtraction method by steadily estimating noise signals in non-speech regions. More >

Jinrong Hu¹, Zhiqin Lei¹, Xiaoying Li^{2, *}, Yongqun He³, Jiliu Zhou¹

CMC-Computers, Materials & Continua, Vol.65, No.1, pp. 705-722, 2020, DOI:10.32604/cmc.2020.09878 - 23 July 2020

Abstract The quality of ultrasound scanning images is usually damaged by speckle noise. This paper proposes a method based on local statistics extracted from a histogram to reduce ultrasound speckle through a region growing algorithm. Unlike single statistical moment-based speckle reduction algorithms, this method adaptively smooths the speckle regions while preserving the margin and tissue structure to achieve high detectability. The criterion of a speckle region is defined by the similarity value obtained by matching the histogram of the current processing window and the reference window derived from the speckle region in advance. Then, according to More >

Jian Wang^{1, 2}, Ming Fang^{3, *}, Hui Li^{1, 2}

CMES-Computer Modeling in Engineering & Sciences, Vol.124, No.1, pp. 79-106, 2020, DOI:10.32604/cmes.2020.09470 - 19 June 2020

Abstract Structural components may enter an initial-elastic state, a plastic-hardening state and a residual-elastic state during strong seismic excitations. In the residual-elastic state, structural components keep in an unloading/reloading stage that is dominated by a tangent stiffness, thus structural components remain residual deformations but behave in an elastic manner. It has a great potential to make model order reduction for such structural components using the tangent-stiffness-based vibration modes as a reduced order basis. In this paper, an adaptive substructure-based model order reduction method is developed to perform nonlinear seismic analysis for structures that have a priori… More >

Jiyuan Zhang¹, Linbo Wang¹, Chengqiang Zhang¹, Shuzhan Bai^2,*

FDMP-Fluid Dynamics & Materials Processing, Vol.16, No.3, pp. 615-621, 2020, DOI:10.32604/fdmp.2020.09654 - 25 May 2020

Abstract The influences of steam and sulfide on the efficiency of NOx reduction using ammonia (NH₃) over the nanometer-class V-W/Ti catalyst in conditions of high temperature is experimentally investigated using a steady-flow reactor. The results showed that selective catalytic reduction (SCR) is inhibited by H₂O at low temperature, but higher NO conversion efficiency is achieved at high temperature since the reaction of NH3 oxidized by O₂ to NOx is inhibited by H₂O. The activity of SCR is promoted by SO₂ in the temperature range of 200~500° C, the NO conversion efficiency was improved to 98% from 94% by adding More >