Kazuhiko Kakuda^1,*, Tomoyuki Enomoto¹, Shinichiro Miura²

CMES-Computer Modeling in Engineering & Sciences, Vol.118, No.1, pp. 1-14, 2019, DOI:10.31614/cmes.2019.04676

Abstract The nonlinear activation functions in the deep CNN (Convolutional Neural Network) based on fluid dynamics are presented. We propose two types of activation functions by applying the so-called parametric softsign to the negative region. We use significantly the well-known TensorFlow as the deep learning framework. The CNN architecture consists of three convolutional layers with the max-pooling and one fully-connected softmax layer. The CNN approaches are applied to three benchmark datasets, namely, MNIST, CIFAR-10, and CIFAR-100. Numerical results demonstrate the workability and the validity of the present approach through comparison with other numerical performances. 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, More >

Tomoyuki Enomoto, Kazuhiko Kakuda, Shinichiro Miura

The International Conference on Computational & Experimental Engineering and Sciences, Vol.21, No.2, pp. 36-39, 2019, DOI:10.32604/icces.2019.05292

Abstract In this paper, the nonlinear activation functions based on fluid dynamics are presented. We propose two types of activation functions by applying the so-called parametric softsign to the negative region. We apply the activation function to CNN (Convolutional Neural Network) which performs image recognition and approaches from multiple benchmark datasets such as MNIST, CIFAR-10. Numerical results demonstrate the workability and the validity of the present approach through comparison with other numerical performances. More >

Jiaxin Wang^1,*, Mengxin Cheng¹, Peng Wu¹, Beijing Chen^1,2

Journal of Information Hiding and Privacy Protection, Vol.1, No.2, pp. 87-93, 2019, DOI:10.32604/jihpp.2019.07189

Abstract Internet brings us not only the convenience of communication but also some security risks, such as intercepting information and stealing information. Therefore, some important information needs to be hidden during communication. Steganography is the most common information hiding technology. This paper provides a literature review on digital image steganography. The existing steganography algorithms are classified into traditional algorithms and deep learning-based algorithms. Moreover, their advantages and weaknesses are pointed out. Finally, further research directions are discussed. More >

Yongxiang Gu¹, Beijing Chen¹, Xu Cheng^1,*, Yifeng Zhang^2,3, Jingang Shi⁴

Journal of Information Hiding and Privacy Protection, Vol.1, No.1, pp. 23-33, 2019, DOI:10.32604/jihpp.2019.05938

Abstract Recently, deep learning has achieved great success in visual tracking. The goal of this paper is to review the state-of-the-art tracking methods based on deep learning. First, we categorize the existing deep learning based trackers into three classes according to network structure, network function and network training. For each categorize, we analyze papers in different categories. Then, we conduct extensive experiments to compare the representative methods on the popular OTB-100, TC-128 and VOT2015 benchmarks. Based on our observations. We conclude that: (1) The usage of the convolutional neural network (CNN) model could significantly improve the… More >

Fangming Bi^1,2, Xin Ma^1,2, Wei Chen^1,2,*, Weidong Fang³, Huayi Chen^1,2, Jingru Li^1,2, Biruk Assefa^1,4

Journal of New Media, Vol.1, No.2, pp. 63-74, 2019, DOI:10.32604/jnm.2019.06253

Abstract Video object tracking is an important research topic of computer vision, which finds a wide range of applications in video surveillance, robotics, human-computer interaction and so on. Although many moving object tracking algorithms have been proposed, there are still many difficulties in the actual tracking process, such as illumination change, occlusion, motion blurring, scale change, self-change and so on. Therefore, the development of object tracking technology is still challenging. The emergence of deep learning theory and method provides a new opportunity for the research of object tracking, and it is also the main theoretical framework More >

Hao Wu¹, Qi Liu^{2, 3, *}, Xiaodong Liu⁴

CMC-Computers, Materials & Continua, Vol.60, No.2, pp. 575-597, 2019, DOI:10.32604/cmc.2019.03595

Abstract Deep learning technology has brought great impetus to artificial intelligence, especially in the fields of image processing, pattern and object recognition in recent years. Present proposed artificial neural networks and optimization skills have effectively achieved large-scale deep learnt neural networks showing better performance with deeper depth and wider width of networks. With the efforts in the present deep learning approaches, factors, e.g., network structures, training methods and training data sets are playing critical roles in improving the performance of networks. In this paper, deep learning models in recent years are summarized and compared with detailed More >

Maoyang Zou^1,2, Jinrong Hu², Huan Zhang², Xi Wu², Jia He², Zhijie Xu³, Yong Zhong^1,*

CMC-Computers, Materials & Continua, Vol.60, No.2, pp. 511-525, 2019, DOI:10.32604/cmc.2019.05912

Abstract For image-guided radiation therapy, radiosurgery, minimally invasive surgery, endoscopy and interventional radiology, one of the important techniques is medical image registration. In our study, we propose a learning-based approach named “FIP-CNNF” for rigid registration of medical image. Firstly, the pixel-level interest points are computed by the full convolution network (FCN) with self-supervise. Secondly, feature detection, descriptor and matching are trained by convolution neural network (CNN). Thirdly, random sample consensus (Ransac) is used to filter outliers, and the transformation parameters are found with the most inliers by iteratively fitting transforms. In addition, we propose “TrFIP-CNNF” which… More >

Hongwei Guo³, Xiaoying Zhuang^3,4,5, Timon Rabczuk^1,2,*

CMC-Computers, Materials & Continua, Vol.59, No.2, pp. 433-456, 2019, DOI:10.32604/cmc.2019.06660

Abstract In this paper, a deep collocation method (DCM) for thin plate bending problems is proposed. This method takes advantage of computational graphs and backpropagation algorithms involved in deep learning. Besides, the proposed DCM is based on a feedforward deep neural network (DNN) and differs from most previous applications of deep learning for mechanical problems. First, batches of randomly distributed collocation points are initially generated inside the domain and along the boundaries. A loss function is built with the aim that the governing partial differential equations (PDEs) of Kirchhoff plate bending problems, and the boundary/initial conditions More >

Cosmin Anitescu¹, Elena Atroshchenko², Naif Alajlan³, Timon Rabczuk^3,*

CMC-Computers, Materials & Continua, Vol.59, No.1, pp. 345-359, 2019, DOI:10.32604/cmc.2019.06641

Abstract We present a method for solving partial differential equations using artificial neural networks and an adaptive collocation strategy. In this procedure, a coarse grid of training points is used at the initial training stages, while more points are added at later stages based on the value of the residual at a larger set of evaluation points. This method increases the robustness of the neural network approximation and can result in significant computational savings, particularly when the solution is non-smooth. Numerical results are presented for benchmark problems for scalar-valued PDEs, namely Poisson and Helmholtz equations, as More >