Mingyang Duan, Jin Liu^*, Shiqi Lv

Journal on Big Data, Vol.3, No.2, pp. 77-83, 2021, DOI:10.32604/jbd.2021.016674

Abstract Image caption generation is an essential task in computer vision and image understanding. Contemporary image caption generation models usually use the encoder-decoder model as the underlying network structure. However, in the traditional Encoder-Decoder architectures, only the global features of the images are extracted, while the local information of the images is not well utilized. This paper proposed an Encoder-Decoder model based on fused features and a novel mechanism for correcting the generated caption text. We use VGG16 and Faster R-CNN to extract global and local features in the encoder first. Then, we train the bidirectional LSTM network with the fused… More >

Shengyang Feng^1,2,3,*, Dongbo Xiong⁴, Guojie Chen⁵, Yu Cui¹, Puxin Chen¹

FDMP-Fluid Dynamics & Materials Processing, Vol.16, No.3, pp. 651-663, 2020, DOI:10.32604/fdmp.2020.07981

Abstract Convection and diffusion are the main factors affecting radon migration. In this paper, a coupled diffusion-convection radon migration model is presented taking into account turbulence effects. In particular, the migration of radon is simulated in the framework of the k-ε turbulence model. The model equations are solved in a complex 3D domain by the finite element method (FEM). Special attention is paid to the case study about radon migration in an abandoned air defense shelter (AADS). The results show that air convection in a confined space has a great influence on the radon migration and the radon concentration is inversely… More >

Senbo Chen^{1, 3, *}, Wenan Tan^{1, 2}

CMC-Computers, Materials & Continua, Vol.64, No.1, pp. 345-358, 2020, DOI:10.32604/cmc.2020.09807

Abstract The problem of influence maximizing in social networks refers to obtaining a set of nodes of a specified size under a specific propagation model so that the aggregation of the node-set in the network has the greatest influence. Up to now, most of the research has tended to focus on monolayer network rather than on multiplex networks. But in the real world, most individuals usually exist in multiplex networks. Multiplex networks are substantially different as compared with those of a monolayer network. In this paper, we integrate the multi-relationship of agents in multiplex networks by considering the existing and relevant… More >

Yanlong Zhang¹, Baoli Yin¹, Yue Cao¹, Yang Liu^{1 , *}, Hong Li¹

CMES-Computer Modeling in Engineering & Sciences, Vol.122, No.3, pp. 1081-1098, 2020, DOI:10.32604/cmes.2020.07822

Abstract In this article, a high-order scheme, which is formulated by combining the quadratic finite element method in space with a second-order time discrete scheme, is developed for looking for the numerical solution of a two-dimensional nonlinear time fractional thermal diffusion model. The time Caputo fractional derivative is approximated by using the L2 -1 _σ formula, the first-order derivative and nonlinear term are discretized by some second-order approximation formulas, and the quadratic finite element is used to approximate the spatial direction. The error accuracy O(h ³ + ∆t ² ) is obtained, which is verified by the numerical results. More >

Xu Zhang¹, Wenpeng Lu^1,*, Fangfang Li^2,3, Xueping Peng³, Ruoyu Zhang¹

CMC-Computers, Materials & Continua, Vol.61, No.2, pp. 601-616, 2019, DOI:10.32604/cmc.2019.06045

Abstract Sentence semantic matching (SSM) is a fundamental research in solving natural language processing tasks such as question answering and machine translation. The latest SSM research benefits from deep learning techniques by incorporating attention mechanism to semantically match given sentences. However, how to fully capture the semantic context without losing significant features for sentence encoding is still a challenge. To address this challenge, we propose a deep feature fusion model and integrate it into the most popular deep learning architecture for sentence matching task. The integrated architecture mainly consists of embedding layer, deep feature fusion layer, matching layer and prediction layer.… More >

J.R. Zabadal¹, B. Bodmann¹, V. G. Ribeiro², A. Silveira², S. Silveira²

CMES-Computer Modeling in Engineering & Sciences, Vol.103, No.4, pp. 215-227, 2014, DOI:10.3970/cmes.2014.103.215

Abstract This work presents a new analytical method to transform exact solutions of linear diffusion equations into exact ones for nonlinear advection-diffusion models. The proposed formulation, based on Bäcklund transformations, is employed to obtain velocity fields for the unsteady two-dimensional Helmholtz equation, starting from analytical solutions of a heat conduction type model. More >

R.H. Haney¹, R.V. Mohan²

CMES-Computer Modeling in Engineering & Sciences, Vol.95, No.5, pp. 431-452, 2013, DOI:10.3970/cmes.2013.095.431

Abstract The use of Graphics Processing Units (GPUs) as co-processors for single CPU/GPU computing systems has become pronounced in high performance computing research, however the solution of truly large scale computationally intensive problems require the utilization of multiple computing nodes. Multiple CPU/GPU computing systems bring new complexities to the observed performance of computationally intensive applications, the more salient of which is the cost of local CPU-GPU host and intra-nodal communication. This paper compares and analyzes the performance of a computationally intensive application represented by resin infusion flow during liquid composite molding process for the manufacture of structural composites application via two… More >

L. Bennamoun¹, A. Belhamri², A. Ali Mohamed

FDMP-Fluid Dynamics & Materials Processing, Vol.5, No.2, pp. 177-192, 2009, DOI:10.3970/fdmp.2009.005.177

Abstract This study focuses on the interplay between drying kinetics (encountered in typical industrial processes and particularly in the context of solar drying) and the possible variation of external (e.g., environmental) conditions. Theoretical models of these behaviours are introduced. Experimental results confirmed by simulation are also presented. Variation of the thermo physical properties of air is taken into account in terms of variation of viscosity, density and coefficient of diffusion. In particular, this coefficient is calculated from experimental data and expressed as a function of the wet bulb air temperature. When external conditions are modified and, as a natural consequence, also… More >

Marco A.A. Kappel¹, Diego C. Knupp¹, Roberto P. Domingos¹, IvanN. Bastos¹

CMC-Computers, Materials & Continua, Vol.49-50, No.1, pp. 13-29, 2015, DOI:10.3970/cmc.2015.049.013

Abstract This work is aimed at the direct and inverse analysis of hydrogen permeation in steels employing a novel anomalous diffusion model. For the inverse analysis, experimental data for hydrogen permeation in a 13% chromium martensitic stainless steel, available in the literature [Turnbull, Carroll and Ferriss (1989)], was employed within the Bayesian framework for inverse problems. The comparison between the predicted values and the available experimental data demonstrates the feasibility of the new model in adequately describing the physical phenomena occurring in this particular problem. More >