• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.97-100
• /
• 2018

The floating holograms currently being commercialized are one-way floating holograms. One-way floating holograms are used for performances and exhibitions, and the implementation method is to reflect the image contents into the reflecting glass at $45^{\circ}$ through a single presentation medium. This creates as inefficient a space as the remaining space of reflective glass tilted at $45^{\circ}$, and even more inefficient space if the video content is presented larger. Therefore, the installation space of a one-way floating hologram is restricted. This paper aims to improve users' satisfaction by developing an upper and lower floating hologram with efficient installation of a one-way floating hologram that is currently commercialized, integrating sound output systems and producing new types of speaker products.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.6B
• /
• pp.553-562
• /
• 2003

Supporting QoS (Quality of Service) for the multimedia applications becomes an important issue as the demand of multimedia applications increases. Thus, it is necessary for the application layer to have an efficient congestion control algorithm, which can support the multimedia applications' QoS requirements. In this paper, we propose a new application layer congestion control algorithm, called RRC-OTT (Receiver-based Rate Control with One-way Trip Time). RRC-OTT algorithm differs from the previously proposed algorithms in that the receiver takes the responsibility of the network congestion control. Thus, RRC-OTT algorithm can not only precisely estimate the network congestion using OTT (one-way Trip Time), but reduce the work load from the sender (e.g., the web server). Our simulation study shows that RRC-OTT algorithm can maintain the comparable link utilization to the previously proposed algorithms and keep the packet jitter low, which thus can help enhance the quality of multimedia applications.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.07a
• /
• pp.51-54
• /
• 2016

In computer vision especially in image processing, it has become popular to apply deep convolutional networks for supervised learning. Convolutional networks have shown a state of the art results in classification, object recognition, detection as well as semantic segmentation. However, supervised learning has two major disadvantages. One is it requires huge amount of labeled data to get high accuracy, the second one is to train so much data takes quite a bit long time. On the other hand, unsupervised learning can handle these problems more cheaper way. In this paper we show efficient way to learn features for classification in an unsupervised way. The network trained layer-wise, used backpropagation and our network learns features from unlabeled data. Our approach shows better results on Caltech-256 and STL-10 dataset.

• International Journal of Ocean Engineering and Technology Speciallssue:Selected Papers
• /
• v.4 no.1
• /
• pp.38-43
• /
• 2001

This paper provides review of research results undertaken on reinforced concrete wall panels in one way and two way. The review also highlights two well accepted code design methods from the American (ACI) and Australia Concrete structures standards. The emphasis is on walls under axial compression only with changes in various parameters. These include the variation of panel dimensions panels (ie. Slenderness, thinness and aspect ratios), steel reinforcement, eccentricities, concrete strength and support conditions. The main purpose of this review is to compile research previous by undertaken to highlight the inadequate in certain research literature. It is envisage that this review will expose areas in wall research required so that inadequate in current methods can be rectified.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.56.2-56
• /
• 2002

A dye process of textile industry is important part that finally gives required quality to the textile material. To produce a proper dye for the customers, color management that precise color measurement from original color sample and rapid estimation of corresponding color recipe is most essential. In the practical dye manufacturing process, color management falls into two categories. First one is color management with an expert knowledge of colorist, the other one is computer aided way. In the former management way, color management fully depends on colorists' expert without support of the measurement and computational devices. Objective color management is impossible in this way. The...

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2004.11a
• /
• pp.251-256
• /
• 2004

The image classification is one of the most important studies in the remote sensing. In general, the MLC(Maximum Likelihood Classification) classification that in consideration of distribution of training information is the most effective way but it produces a bad result when we apply it to actual hyperspectral image with the same classification technique. The purpose of this research is to reveal that which one is the most effective and suitable way of the classification algorithms iii the hyperspectral image classification. To confirm this matter, we apply the MLC classification algorithm which has distribution information and SAM(Spectral Angle Mapper), SFF(Spectral Feature Fitting) algorithm which use average information of the training class to both multispectral image and hyperspectral image. I conclude this result through quantitative and visual analysis using confusion matrix could confirm that SAM and SFF algorithm using of spectral pattern in vector domain is more effective way in the hyperspectral image classification than MLC which considered distribution.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.19 no.6
• /
• pp.91-95
• /
• 1982

One significant problem which arises during the fabrication of LSI and/or VLSI is how to make a good step coverage for aluminum. One way developed in this study is to flow PSG containing 8.0 wt.% - 8.6 wt.% of phosphorus at 95$0^{\circ}C$ for 30 minutes in steam and to depo3it aluminum of 1 urn thickness at $250^{\circ}C$.

• Journal of the Korean Statistical Society
• /
• v.15 no.1
• /
• pp.46-61
• /
• 1986

Many procedures for deriving the Moore-Penrose invese $X^+$ have been developed, but the explicit forms of Moore-Penerose inverses for design matrices in analysis of variance models are not known heretofore. The purpose of this paper is to find explicit forms of $X^+$ for the one-way and the two-way analysis of variance models. Consequently, the Moore-Penerose inverse $X^+$ and the shortest solutions of them can be easily obtained to the level of pocket calculator by way of our results.

• Journal of Korean Society for Quality Management
• /
• v.10 no.1
• /
• pp.7-12
• /
• 1982

Arvesen (1969) has shown a procedure which produces an approximate confidence interval for a variance component in unbalanced one-way classification model. In this paper, his work is extended to the case when the model of interest is unbalanced two-way classification. Following the procedure described in this paper, approximate confidence intervals are computed by a Monte Carlo simulation.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.133-137
• /
• 2002

This paper proposes a way of constructing modern concept of geographic network by interpreting historical geography recorded in old archives such as the Tongdian(通典, Comprehensive Manual) which is one of the important materials in studying old concept of Chinese administrative geography. it is proposed a way of producing geographic network by constructing base network, and by combining and adjusting the constructed networks. The proposed way of construction has been applied for generating a geographic network around Bohai gulf and Sandong peninsula, by using descriptions recorded in the administrative division(州郡典) of the Tongdian.