• Transactions of the Korean Society of Automotive Engineers
• /
• v.23 no.1
• /
• pp.1-10
• /
• 2015

In urban area, since multipath and signal attenuations frequently occur due to street trees, street lights and buildings, it is difficult to obtain accurate navigation solution using GPS. As these problems also impact negatively on the INS/GPS coupled system, implementing advanced transportation systems such as autonomous navigation system and Intelligent Transportation System (ITS) become quite hard. For this reason, to alleviate deterioration of navigation system performance in urban area, direction information extraction algorithm using vision system is proposed in this paper. 2D Finite Rate of Innovation (FRI) technique is applied to extract lane edges. The proposed technique is simulated using road images and feasibility of proposed technique is analyzed through the simulation results.

• Journal of Fashion Business
• /
• v.14 no.4
• /
• pp.146-160
• /
• 2010

The influence of mass media on the consumers who want to purchase fashion products might be very important. Especially fashion magazine, a media of providing professional information and opinion about fashion, not only provides the information of products or fashion trend, but is used as an advertising media for publicizing products. The purpose of this study is to investigate how such fashion magazine affects consumers. In this study, questionnaire survey was administrated to understand the status of subscription of fashion magazine and consumers' influence. To analyze the garment style comparatively, the pictures that were taken on Dongseong Street from December 2007 to November 2008, and the pictures in fashion magazine Vogue Korea were analyzed to the content. Followings are the results of the study. consumers receive the information of products and new style through fashion magazine, but mostly wear the garment style that emphasizes convenience in practical life. In other words, the fashion magazine provides consumers with the information of fashion products and has the influence as the source of fashion style information and the fashion trend, but is not yet spreaded to people as much as generalized and accepted on the street. It is because the information in fashion magazine is differentiated information that is first received by particular readers, that is, the fashion leaders sensitive and much interested in the fashion. Fashion magazine is used as a means of getting information of fashion and style and satisfying the desire of novelty, and tend to be a sensuous media of delivering information through image. It, however, must be a mass media that shows the trend and leads the fashion.

• Journal of the Korean Society of Costume
• /
• v.62 no.2
• /
• pp.103-121
• /
• 2012

The purpose of this study is to provide basic information to improve competitiveness of Korean fashion brands in the Chinese fashion market by taking photos of street styles in Seoul that is the hub of Korean fashion and in Dalian, the northeastern district where Korean fashion brands are launched as a test market China to compare and analyze the clothing colors preferred mostly by young women in their twenties and thirties who have the highest purchasing power. The study methods used literature review and empirical study simultaneously. Dalian and Seoul are two fashion cities in northeast of China and Korea where street fashion was photographed. Clothing colors that were mainly worn by young women in their twenties and thirties were qualitatively analyzed using the photos taken. Color analysis was based on the Munsell Color Order System to grasp general preferences of colors on the basis of previous researches, and color tones were based on the ISCC-NBS System. In order to grasp the basic materials on Seoul, the Korean fashion city and Dalian, the northeastern fashion city in China, street fashion styles of 2010/11 F/W season were compared and analyzed. As a result, black and deep tone PB color appeared most frequently. The vivid and strong tone of R, YR, Y color showed high frequency of clothing colors. For the top wear, women in both areas preferred similar tones in the YR color category and Dalian women preferred vivider and brighter S tones in the R color category. For the bottom wear, women in both areas highly preferred achromatic colors and colors in the PB color category. For bags, women in both areas preferred black and colors in the YR color category. Finally, for the shoes, while women in both areas preferred black in the achromatic color category, they showed different color preferences in the chromatic color category. R color categories were preferred by the women in Seoul and the YR color categories were preferred by the women in Dalian. Conclusively, women in both cities highly preferred achromatic colors especially black for the top and bottom wear, bags, and shoes. It may reflect their desire to look refined and slender through colors. Moreover, continuous and steady research on fashion trends in Seoul and Dalian may have positive effects on Korean fashion businesses that plan to be launched in China, the area of competition for global brands.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.35 no.5
• /
• pp.431-438
• /
• 2017

The information management system for urban planning, urban management and related decision making is increasing an importance in recent years. Especially, the information about 3D buildings in urban area is necessary for urban management related precise location based service. However, it takes a lot of time, effort, and costs to produce 3D modeling data of buildings. These problems have been found in many studies, but the research to solve these problems is still necessary. In this study, we have tested an efficient 3D modeling method of building to solve these problems. SketchUp was used to test 3D modeling methods with OSM(OpenStreetMap) data. The other 3D modeling method using a digital map was tested. The 3D modeling of building data was efficient using PlaceMaker, which can use OSM data. In addition, modeling was possible through simple editing of the digital map and automatic building generation. The efficiency of the research results is presented by comparing with the traditional method based on the regulation and the stadard man power issued by Korean government. These two 3D modeling methods of building tested through the study can be used as basic data for related research and work. Additional studies should be conducted to improve the accuracy of the building model and determine how to obtain height data efficiently.

• Proceedings of the Korea Contents Association Conference
• /
• 2004.11a
• /
• pp.501-504
• /
• 2004

Straight traffic signal information is understanded by roation traffic signal.Ten traffic rotation street flew system saturated without forseed quantaty of traffic vehicle and length oftraffic street.Optimul traffic signal forcasting is needed departmemtstore sale season. for solution of this problem, at least before 10 hour traffic quantity reported policy department.For traffic forcasting of traffic jamed building pushed vehicle, neural network method is adapted Single traffic network is more needed to through traffic signal compared of traffic network. traffic network system is first analysed for artificial traffic network system and networked traffic systen is constructed by ubiquatous.

• Proceedings of the Korea Contents Association Conference
• /
• 2016.05a
• /
• pp.239-240
• /
• 2016

Traditional town with wooden houses like Japanese historic local town is vulnerable to disaster. Wooden houses crowed in this study area, there is no gap between buildings and buildings. When disaster occurs, they can not use evacuation routes based on the value of probability of rubble flow. Disaster prevention design of traditional town using spatial information for residents is a modern idea in the field of disaster prevention study. Therefore, it is basically important and effective to create information, especially on the current situation such as dangerousness of the area. Here, we report on a disaster prevention design regarding probability of street blockade and probability of rubble flow at a large earthquake directly under our study area. In this study, we explain about necessity of temporary evacuation areas like open space. As a result, we found some ideas to secure evacuation routes from traditional houses to designated refuge places using temporary evacuation areas.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.5A
• /
• pp.517-525
• /
• 2008

In this paper, we cary out system level simulations to investigate the effect of cell shape(i.e., different base station displacements in the two directions defined by the street grid) on minimizing transmitter power, interference power, and blocking probability for CDMA system in urban microcellular environments. In urban microcell, path loss to the base station depends on the orientation of the street where the mobile is located. Interference from mobile stations to the base station in the reference cell is considered up to second tier. The wrap around method is used to include the second tier interference with realistic computational complexity without reducing the accuracy of interference calculations. The investigation shows that the transmitter power, interference power, and blocking probability in a cell can be reduced by proper selection of the efficient cell shape.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.515-518
• /
• 2019

Semantic segmentation has remained as a challenging problem in the field of computer vision. Given the immense power of Convolution Neural Network (CNN) models, many complex problems have been solved in computer vision. Semantic segmentation is the challenge of classifying several pixels of an image into one category. With the help of convolution neural networks, we have witnessed prolific results over the time. We propose a convolutional neural network model which uses Fully CNN with deconvolutional pixel layers. The goal is to create a hierarchy of features while the fully convolutional model does the primary learning and later deconvolutional model visually segments the target image. The proposed approach creates a direct link among the several adjacent pixels in the resulting feature maps. It also preserves the spatial features such as corners and edges in images and hence adding more accuracy to the resulting outputs. We test our algorithm on Karlsruhe Institute of Technology and Toyota Technologies Institute (KITTI) street view data set. Our method achieves an mIoU accuracy of 92.04 %.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.1
• /
• pp.183-189
• /
• 2011

This paper suggests a user based OST(Open Street Traffic) system that solves TPEG's one-way communication problem, UTIS's limited usage and DSRC's small traffic bandwidth. In current commercial TPEG service, only some service providers collect traffic information. Thus, it can't cover traffic status in local lanes And UTIS, which local governments and police supports, requires additional equipments. Currently, only taxi and official vehicles use this system. Therefore, new traffic service by mobile device and user's participation can provide very detail traffic information coupling with previous traffic systems. But in this new system, real-time high volume data can be a problem. So, in this paper, new data storage manager design(TDSM :Traffic Data Storage Manager) is suggested and its performance is measured against commercial DBMS.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.05a
• /
• pp.94-95
• /
• 2006

A great deal of research has been made to model the vagueness and uncertainty in information retrieval. One such research is fuzzy ranking models, which have been showing their superior performance in handling the uncertainty involved in the retrieval process. In this study we develop a new fuzzy ranking model based on the user preference. Through the experiments on the TREC-2 collection of Wall Street Journal documents, we show that the proposed method outperforms the conventional fuzzy ranking models.