• Journal of Korean Tunnelling and Underground Space Association
• /
• v.11 no.1
• /
• pp.37-45
• /
• 2009

In this study, experiments of development and application of 50 MPa high-performance concrete are performed for large dimensional tunnel linings. In order to produce 50MPa high-performance concrete, eight optimal mixtures replacing with fly ash and ground granulated blast furnace slag up to 50 percent of type I Portland cement were selected then tests for mechanical properties and simple adiabatic temperature rise tests were carried out. And in order to assess the quantitative characteristics of heat of hydrations of developed mixtures, three mixtures that the type I Portland cement (OPC) and each one mixture of binary and ternary mixtures (BS30, F15S35) were reselected, then the adiabatic temperature rise tests and mock-up tests were performed. Consequently, the comparisons between the results of mock-up tests and finite element analysis can be enhanced the reliability of analyzing routines of thermal behaviours of the developed high-performance concrete.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.54-63
• /
• 2011

This paper proposes a fast learning-based interpolation algorithm to up-scale an input low-resolution image into a high-resolution image. In conventional learning-based super-resolution, a certain relationship between low-resolution and high-resolution images is learned from various training images and a specific high frequency synthesis information is derived. And then, an arbitrary low resolution image can be super-resolved using the high frequency synthesis information. However, such super-resolution algorithms require heavy memory space to store huge synthesis information as well as significant computation due to two-dimensional matching process. In order to mitigate this problem, this paper presents one-dimensional patch-based learning and synthesis. So, we can noticeably reduce memory cost and computational complexity. Simulation results show that the proposed algorithm provides higher PSNR and SSIM of about 0.7dB and 0.01 on average, respectively than conventional bicubic interpolation algorithm.

• Proceedings of the KAIS Fall Conference
• /
• 2003.11a
• /
• pp.223-230
• /
• 2003

Feature-based similarity retrieval has become an important research issue in multimedia database systems. The features of multimedia data are useful for discriminating between multimedia objects (e 'g', documents, images, video, music score, etc.). For example, images are represented by their color histograms, texture vectors, and shape descriptors, and are usually high-dimensional data. The performance of conventional multidimensional data structures(e'g', R- Tree family, K-D-B tree, grid file, TV-tree) tends to deteriorate as the number of dimensions of feature vectors increases. The R＊-tree is the most successful variant of the R-tree. In this paper, we propose a SOM-based R＊-tree as a new indexing method for high-dimensional feature vectors.The SOM-based R＊-tree combines SOM and R＊-tree to achieve search performance more scalable to high dimensionalities. Self-Organizing Maps (SOMs) provide mapping from high-dimensional feature vectors onto a two dimensional space. The mapping preserves the topology of the feature vectors. The map is called a topological of the feature map, and preserves the mutual relationship (similarity) in the feature spaces of input data, clustering mutually similar feature vectors in neighboring nodes. Each node of the topological feature map holds a codebook vector. A best-matching-image-list. (BMIL) holds similar images that are closest to each codebook vector. In a topological feature map, there are empty nodes in which no image is classified. When we build an R＊-tree, we use codebook vectors of topological feature map which eliminates the empty nodes that cause unnecessary disk access and degrade retrieval performance. We experimentally compare the retrieval time cost of a SOM-based R＊-tree with that of an SOM and an R＊-tree using color feature vectors extracted from 40, 000 images. The result show that the SOM-based R＊-tree outperforms both the SOM and R＊-tree due to the reduction of the number of nodes required to build R＊-tree and retrieval time cost.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.121-124
• /
• 2007

This paper describes the development of a brickwork game with three-dimensional computer graphics as one of web-based game contents. Client user using web can access and run lava applet program with the independence of hardware system. It consists of graphic user interface module, initial space generation module, event handler module, player control module, and thread control module. It uses 3-D array data structure for the 3-D graphic objects that are located in three-dimensional space for high-speed object searching and sorting. It enhances to compare with predetermined construction in three-dimensional space. We can use the developed racing game to inform game users of information for an advertisement like tourism information, and can apply the proposed 3-D drawing technology to 3-D game graphic engine core.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.6
• /
• pp.15-20
• /
• 2015

Bit patterned media (BPM) is a high capacity storage system and has attracted a great deal of attention as next generation data storage. When BPM is made with high density, the space between the islands narrows, because BPM records one bit in an island. For this reason, BPM has inter-symbol interference in all directions, unlike in current storage systems where it is in only one direction. In this paper, we propose an equation for compensating two-dimensional ISI. We conduct experiments on track misregistration. When using the proposed inter-symbol interference preprocessing, the BER performance is improved, regardless of the amount of track misregistration.

• The Korean Journal of Applied Statistics
• /
• v.36 no.2
• /
• pp.167-173
• /
• 2023

This paper investigates several methods of visualizing high-dimensional data in a low-dimensional space. At first, principal component analysis and multidimensional scaling are briefly introduced as linear approaches, and then kernel principal component analysis, self-organizing map, locally linear embedding, Isomap, Laplacian Eigenmaps, and local multidimensional scaling are introduced as nonlinear approaches. In particular, t-SNE, which is widely used but relatively unfamiliar in the field of statistics, is described in more detail. We also present a simple example for several methods, including t-SNE. Finally, we provide a review of several recent studies pointing out the limitations of t-SNE and discuss the future research problems presented.

• Bulletin of the Korean Space Science Society
• /
• 2003.10a
• /
• pp.82-82
• /
• 2003

It is well known that electromagnetic (EM) waves are mode converted to electrostatic (ES) waves in inhomogeneous plasmas. We examine this issue in a three-dimensional multi-fluid numerical model. First, we derive a set of coupled linear wave equations when a one-dimensional inhomogeneous density profile is assumed in a cold and collisionless plasma. The massive ions are considered as fixed because we are interested in high frequency waves in plasmas. It is shown that the EM mode satisfies the 0th order modified Bessel equation near the resonant region where the frequency matches the local electron plasma frequency. It is expected that the EM waves are coupled and damped to the ES waves owing to the logarithmic singular behavior at such resonances. Second, we numerically test the same case in a 3-D multi-fluid model. An impulsive input is assumed to excite EM waves in the inhomogeneous 3-D box model. The wave spectra of electric and magnetic fields are presented and compared with the analytical results. Our results suggest that the EM energy is irreversibly converted into the ES energy wherever the resonant condition is satisfied. Finally we discuss how the mode conversion appears in both electric and magnetic fields by analyzing time histories of each component. We also compare our results with MHD wave coupling. It is numerically confirmed in this study that the coupling of EM and ES waves is similar to that of compressional and transverse MHD waves.

• Journal of Korean Society for Quality Management
• /
• v.48 no.1
• /
• pp.51-68
• /
• 2020

Purpose: The primary objective of this research is to develop the optimal operating conditions as well as their associated design spaces for a Cryogenic Submerged Arc Welding(SAW) process by improving its quality and productivity simultaneously. Methods: In order to investigate functional relationships among quality characteristics and their associated control factors of an SAW process, a stepwise design of experiment(DoE) method is proposed in this paper. Based on the DoE results, not only a multi-dimensional design space but also a safe operating space and normal acceptable range(NAR) by integrating statistical confidence intervals were demonstrated. In addition, the optimal operating conditions within the proposed NAR can be obtained by a robust optimal design method. Results: This study provides a customized stepwise DoE method (i.e., a sequential set of DoE such as a factorial design and a central composite design) for Cryogenic SAW process and its statistical analysis results. DoE results can then provide both the main and interaction effects of input control factors and the functional relationships between the input factors and their associated output responses. Maximizing both the product quality with high impact strength and the productivity with minimum processing times simultaneously in a case study, we proposed a design space which can provide both acceptable productivity and quality levels and NARs of input control factors. In order to confirm the optimal factor settings and the proposed NARs, validation experiments were performed. Conclusion: This research may provide significant contributions and applications to many SAW problems by preparing a standardization of the functional relationship between the input factors and their associated output response. Moreover, the proposed design space based on DoE and NAR methods can simultaneously consider a number of quality characteristics including tradeoff between productivity and quality levels.

• BMB Reports
• /
• v.37 no.1
• /
• pp.28-34
• /
• 2004

The discovery of biochemical and cellular functions of unannotated gene products begins with a database search of proteins with structure/sequence homologues based on known genes. Very recently, a number of frontier groups in structural biology proposed a new paradigm to predict biological functions of an unknown protein on the basis of its three-dimensional structure on a genomic scale. Structural proteomics (genomics), a research area for structure-based functional discovery, aims to complete the protein-folding universe of all gene products in a cell. It would lead us to a complete understanding of a living organism from protein structure. Two major complementary experimental techniques, X-ray crystallography and NMR spectroscopy, combined with recently developed high throughput methods have played a central role in structural proteomics research; however, an integration of these methodologies together with comparative modeling and electron microscopy would speed up the goal for completing a full dictionary of protein folding space in the near future.

• The Bulletin of The Korean Astronomical Society
• /
• v.43 no.2
• /
• pp.38.3-38.3
• /
• 2018

We present the results of high-resolution near-IR spectral mapping toward the Orion KL outflow. In this study, we used the Immersion Grating Infrared Spectrometer (IGRINS) on the 2.7 m Harlan J. Smith Telescope at McDonald Observatory. IGRINS's large wavelength coverage over the H & K bands and high spectral resolving power (R ~ 45,000) allowed us to detect over 35 shock-excited ro-vibrational H2 transitions and to measure directly the gas temperature and velocity of the dense outflows. In our previous study toward the H2 peak 1 region in the Orion KL outflow, we identified 31 outflow fingers from a datacube of the H2 1-0 S(1) $2.122{\mu}m$ line and constructed a three-dimensional map of the fingers. The internal extinction (${\Delta}AV$ > 10 mag) and overall angular spread of the flow argue for an ambient medium with a high density (105 cm-3). In this presentation, we show preliminary results of additional mapping toward a remarkable chain of bows (HH 205 - HH 207) farther from the ejection center, and obtain a more clear view of the shock physics of a single isolated bullet that improves on the knowledge gained from observations of the more complex peak 1 region in our earlier study.