• Journal of Information Processing Systems
• /
• v.14 no.3
• /
• pp.590-599
• /
• 2018

Multivariate finite mixture model is becoming more and more popular in image processing. Performing image denoising from image patches to the whole image has been widely studied and applied. However, there remains a problem that the structure information is always ignored when transforming the patch into the vector form. In this paper, we study the operator which extracts patches from image and then transforms them to the vector form. Then, we find that some pixels which should be continuous in the image patches are discontinuous in the vector. Due to the poor anti-noise and the loss of structure information, we propose a new operator which may keep more information when extracting image patches. We compare the new operator with the old one by performing image denoising in Expected Patch Log Likelihood (EPLL) method, and we obtain better results in both visual effect and the value of PSNR.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.28 no.3
• /
• pp.108-119
• /
• 2024

The Reynolds-averaged Navier-Stokes (RANS) simulations are commonly used in industrial applications due to their computational efficiency. However, the linear eddy viscosity model (LEVM) used in RANS often fails to accurately capture the anisotropy of Reynolds stress in complex flow conditions. To enhance RANS predictive accuracy, data-driven closure models, such as Tensor Basis Neural Network (TBNN) and Tensor Basis Random Forest (TBRF), have been proposed. However existing models, including TBNN and TBRF, have limitations in capturing the nonlocal patterns of turbulence models, resulting in irregular and unsmooth predictions. Convolutional neural networks (CNNs) are considered as an alternative approach, but their reliance on discretization poses challenges when dealing with arbitrarily designed meshes in RANS simulations. In this study, we propose a nonlinear convolutional neural operator as the RANS closure model. Our model satisfies Galilean invariance, can learn nonlocal physics, and recovers high-resolution physics even when trained on undersampled grids. The model outperforms existing TBNN and TBRF models, successfully predicting smooth fields of Reynolds stress in flows with adverse pressure gradients, separations, and streamline curvature, where existing models struggle or fail to provide accurate predictions.

• Information and Communications Magazine
• /
• v.33 no.6
• /
• pp.18-26
• /
• 2016

Network and network functions virtualization (NFV) promise a number of attractive benefits and thus have driven mobile network operators to transform their previously static networks to more dynamic and software-defined networks. In this article, we share a mobile network operator's view based on implementation and deployment experiences in the wild during the past few years towards a software-defined 5G core network. More specifically, we present a practical point of view from mobile network operators and elaborate on why some of the virtualization benefits such as total cost of ownership (TCO) reduction are not easily realized as initially intended. Then, we describe 5G visions, services, and their requirements commonly agreed across mobile operators globally. Given the requirements, we then introduce desirable characteristics of 5G mobile core network and its key enabling technologies.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.496-499
• /
• 1999

This paper describes a medular neural network(MNN) for a vision system which tracks a given object using a sequence of images from a camera unit. The MNN is used to precisely recognize the given robot endeffector and to minize the processing time. Since the robot endeffector can be viewed in many different shapes in 3-D space, a MNN structure, which contains a set of feedforwared neural networks, co be more attractive in recognizing the given object. Each single neural network learns the endeffector with a cluster of training patterns. The training patterns for a neural network share the similar charateristics so that they can be easily trained. The trained MNN is less sensitive to noise and it shows the better performance in recognizing the endeffector. The recognition rate of MNN is enhanced by 14% over the single neural network. A vision system with the MNN can precisely recognize the endeffector and place it at the center of a display for a remote operator.

• ETRI Journal
• /
• v.43 no.1
• /
• pp.152-162
• /
• 2021

In the last few years, detection has become a powerful methodology for network protection and security. This paper presents a new detection scheme for data recorded over a computer network. This approach is applicable to the broad scientific field of information security, including intrusion detection and prevention. The proposed method employs bidimensional (time-frequency) data representations of the forms of the short-time Fourier transform, as well as the Wigner distribution. Moreover, the method applies matrix factorization using singular value decomposition and principal component analysis of the two-dimensional data representation matrices to detect intrusions. The current scheme was evaluated using numerous tests on network activities, which were recorded and presented in the KDD-NSL and UNSW-NB15 datasets. The efficiency and robustness of the technique have been experimentally proved.

• Proceedings of the KIEE Conference
• /
• 1999.07b
• /
• pp.963-965
• /
• 1999

A linear chromosome combined with a grid-based representation of the network and a new crossover operator allow the evolution of the architecture and the weights simultaneously. In our approach there is no need for a separate weight optimization procedure and networks with more than one type of activation function can be evolved. In this paper one evolutionary' strategy of a given dual neural controller was introduced and the simulation results were described in detail through applications to a stabilization control of an Inverted Pendulum System.

• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.637-640
• /
• 1999

In this Paper we show the LCD simulator for defect inspection using image processing algorithm and neural network. The defect inspection algorithm of the LCD consists of preprocessing, feature extraction and defect classification. Preprocess removes noise from LCD image, using morphology operator and neural network is used for the defect classification. Sample images with scratch, pinhole, and spot from real LCD color filter image are used. The proposed algorithms show that defect detected and classified in the ratio of 92.3% and 94.6 respectively.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1996.10a
• /
• pp.223-226
• /
• 1996

One of the consideration issue in implementation and maintanence of CIM(Computer Integrated Manufacturing) database is exchange and sharing of information between heterogeneous databases. For efficient operating of SIM systems, it must be able to organize and to manage the information. In this paper, we propose method that can make enhance the efficiency of CIM database, classfying the data in database using self organize neural network to each database systems, and computing between classfied heterogeneous database using extended operator that is defined.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.45-51
• /
• 1995

A novel definition for fuzzy mathematical morphology is described The generalized-mean operator plays the key role for this definition. Several hard constraints for standard generalized-mean have been eliminated. Complete mathematical description for obtaining fuzzy erosion and dilation is provided. The definitions are well suited for neural network implementation. Therefore, the parameters for the fuzzy definition can be optimized using neural network learning paradigm.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06a
• /
• pp.15-16
• /
• 2007