• Korean Journal of Ecology and Environment
• /
• v.39 no.3 s.117
• /
• pp.352-365
• /
• 2006

In order to evaluate the change of aquatic environment in the reclaimed Hwaong Reservoir, situated in the early stage after construction, this study was conducted to measure the change of precipitation, temperature, and salinity from June 2002 to January 2006. The range and mean of temperature was $-0.7{\sim}33.4^{\circ}C$ and $13.6^{\circ}C$, respectively. Temperature of upstream part rapidly changed during the transitional period; from spring to summer and from fall to winter. It showed abrupt decrease with high discharge from the streams temporarily. While, hypolimnetic temperature of upstream happened to be somewhat higher than that of surface or downstream. The range and mean of salinity was 0.3${\sim}$32.3 psu and 25.3 psu, respectively. Vertical difference of salinity was marked, and the change in the surface water was much higher than middle or bottom layers. It showed the marked difference at all stations, except for the bottom layer of upstream into which Namyang Stream flows, indicating that vertical gradient of salinity is strongly sustained in the reservoir. Salinity was changed markedly during the storm period (June${\sim}$October), and freshwater with low salinity was expanded from upstream to downstream along the surface layer. The surface of the reservoir was totally covered by the stream discharged water with a large amount of silt and low salinity during this period. The difference of temperature and salinity between the surface and bottom layer ranged $-10.6{\sim}9.7^{\circ}C$ and $-27.1{\sim}30.0$ psu, respectively. The big difference of salinity appeared with a large discharge of freshwater from the streams or large input of seawater through the gate. Salinity was negatively correlated with temperature, indicating the influence of monsoon storm events on the salinity under the whole watershed scale of this brackish reclaimed reservoir.

• Journal of Intelligence and Information Systems
• /
• v.24 no.3
• /
• pp.1-19
• /
• 2018

Large amount of data is now available for research and business sectors to extract knowledge from it. This data can be in the form of unstructured data such as audio, text, and image data and can be analyzed by deep learning methodology. Deep learning is now widely used for various estimation, classification, and prediction problems. Especially, fashion business adopts deep learning techniques for apparel recognition, apparel search and retrieval engine, and automatic product recommendation. The core model of these applications is the image classification using Convolutional Neural Networks (CNN). CNN is made up of neurons which learn parameters such as weights while inputs come through and reach outputs. CNN has layer structure which is best suited for image classification as it is comprised of convolutional layer for generating feature maps, pooling layer for reducing the dimensionality of feature maps, and fully-connected layer for classifying the extracted features. However, most of the classification models have been trained using online product image, which is taken under controlled situation such as apparel image itself or professional model wearing apparel. This image may not be an effective way to train the classification model considering the situation when one might want to classify street fashion image or walking image, which is taken in uncontrolled situation and involves people's movement and unexpected pose. Therefore, we propose to train the model with runway apparel image dataset which captures mobility. This will allow the classification model to be trained with far more variable data and enhance the adaptation with diverse query image. To achieve both convergence and generalization of the model, we apply Transfer Learning on our training network. As Transfer Learning in CNN is composed of pre-training and fine-tuning stages, we divide the training step into two. First, we pre-train our architecture with large-scale dataset, ImageNet dataset, which consists of 1.2 million images with 1000 categories including animals, plants, activities, materials, instrumentations, scenes, and foods. We use GoogLeNet for our main architecture as it has achieved great accuracy with efficiency in ImageNet Large Scale Visual Recognition Challenge (ILSVRC). Second, we fine-tune the network with our own runway image dataset. For the runway image dataset, we could not find any previously and publicly made dataset, so we collect the dataset from Google Image Search attaining 2426 images of 32 major fashion brands including Anna Molinari, Balenciaga, Balmain, Brioni, Burberry, Celine, Chanel, Chloe, Christian Dior, Cividini, Dolce and Gabbana, Emilio Pucci, Ermenegildo, Fendi, Giuliana Teso, Gucci, Issey Miyake, Kenzo, Leonard, Louis Vuitton, Marc Jacobs, Marni, Max Mara, Missoni, Moschino, Ralph Lauren, Roberto Cavalli, Sonia Rykiel, Stella McCartney, Valentino, Versace, and Yve Saint Laurent. We perform 10-folded experiments to consider the random generation of training data, and our proposed model has achieved accuracy of 67.2% on final test. Our research suggests several advantages over previous related studies as to our best knowledge, there haven't been any previous studies which trained the network for apparel image classification based on runway image dataset. We suggest the idea of training model with image capturing all the possible postures, which is denoted as mobility, by using our own runway apparel image dataset. Moreover, by applying Transfer Learning and using checkpoint and parameters provided by Tensorflow Slim, we could save time spent on training the classification model as taking 6 minutes per experiment to train the classifier. This model can be used in many business applications where the query image can be runway image, product image, or street fashion image. To be specific, runway query image can be used for mobile application service during fashion week to facilitate brand search, street style query image can be classified during fashion editorial task to classify and label the brand or style, and website query image can be processed by e-commerce multi-complex service providing item information or recommending similar item.

• Journal of the Economic Geographical Society of Korea
• /
• v.10 no.4
• /
• pp.427-443
• /
• 2007

The purpose of this study is to investigate spacio-temporal structures of traffic flows on the subway network in the Metropolitan Seoul, and the relationships between topological structures of traffic flows and land use patterns. In particular we analyze in the topological structures of traffic flows on the subway network in time dimension as well as in spatial dimension. For the purpose, this study utilizes data mining techniques to the one day T-card transaction data of the last four years, which has developed for exploring the characteristics of traffic flows from large scale trip-transaction databases. The topological structures of traffic flows on the subway network has changed considerably during the last four years. The volumes of traffic flows, the travel time and stops per trip have increased until 2006 and decreased again in 2007. The results are visualized by utilizing GIS and analyzed, and thus the spatial patterns of traffic flows are analyzed. The spatial distribution patterns of trip origins and destinations show substantial differences among time zones during a day. We analyze the relationships between traffic flows at subway stops and the geographical variables reflecting land use around them. We obtain 6 log-linear functions from stepwise multiple regression analysis. We test multicollinearity among the variables and autocollelation for the residuals.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.9
• /
• pp.1-10
• /
• 2014

In this paper, we propose MPI gridding algorithm of LiDAR data that minimizes the communication between the cores. The LiDAR data collected from aircraft is a 3D spatial information which is used in various applications. Since there are many cases where the LiDAR data has too high resolution than actually required or non-surface information is included in the data, filtering the raw LiDAR data is required. In order to use the filtered data, the interpolation using the data structure to search adjacent locations is conducted to reconstruct the data. Since the processing time of LiDAR data is directly proportional to the size of it, there have been many studies on the high performance parallel processing system using MPI. However, previously proposed methods in parallel approach possess possible performance degradations such as imbalanced data size among cores or communication overhead for resolving boundary condition inconsistency. We conduct empirical experiments to verify the effectiveness of our proposed algorithm. The results show that the total execution time of the proposed method decreased up to 4.2 times than that of the conventional method on heterogeneous clusters.

• The Journal of the Acoustical Society of Korea
• /
• v.35 no.3
• /
• pp.223-233
• /
• 2016

Design of a vegetation type sound barrier was presented as a noise barrier on the boundary of neighborhood facilities including schools, and apartments. The suggested noise barrier is made of unit blocks that are to be formed by stacking over the wall structure containing the plant and soils in the blocks. The advantage of the vegetation noise barrier is to acquire not only sound absorptive effects of plants and soils, but also sound diffusive effect caused by the irregular surface of the barrier which could eventually mitigate the noise. First, the optimum size of the units to obtain the highest noise reduction was investigated using 1/10 scaled model experiment, and sound attenuation experiments were carried out using a 1/2 mock-up model which is 2 m high and 5 m long. Total 1,137 unit blocks were made of synthetic woods with the size of $10{\times}10{\times}9cm$. These unit blocks were installed on the both side of the 1/2 mock-up steel framed noise barrier. As a result, it was revealed that the block typed vegetation noise barrier has 7 dB higher insertion loss in comparison with the general plane noise barrier. Also, it was found that the appropriate size of unit blocks is $20{\times}20cm$ which has large effect of sound insertion loss.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.35 no.1
• /
• pp.1-8
• /
• 2015

Wind power systems have gained much attention due to the relatively high reliability, good infrastructures and cost competitiveness to the fossil fuels. Advances have been made to increase the power efficiency of wind turbines while less attention has been focused on structural integrity assessment of structural sub-systems such as towers and foundations. Among many parameters for integrity assessment, the most perceptive parameter may be the induced horizontal displacement at the hub height although it is very difficult to measure particularly in large-scale and high-rise wind turbine structures. This study proposes an indirect displacement estimation scheme based on the combined use of inclinometers and accelerometers for more convenient and cost-effective measurements. To this end, (1) the formulation for data fusion of inclination and acceleration responses was presented and (2) the proposed method was numerically validated on an NREL 5 MW wind turbine model. The numerical analysis was carried out to investigate the performance of the propose method according to the number of sensors, the resolution and the available sampling rate of the inclinometers to be used.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.8 no.4
• /
• pp.102-113
• /
• 2005

Ecosystem is composed of human, biotic and abiotic environment. Landscape is an ecosystem which appear in a unit region. These landscape are the spatiotemporal land mosaic which is combined with various landscape elements. And, land use and land cover changes are important factors of landscape structure changes. This study is mainly focused on the analysing the spatiotemporal change patterns of Daegu metropolitan sphere forest landscape, using landscape indices and Ecosystem Service Value (ESV) which quantify ecosystem structures and functions. The results of this study are as follow: The encroachment and fragmentation of forest were due to linear developments, i. e. road construction, rather than large-scale developments such as residental lands or industrial complexes. And, the core area percentages of landscape gradually decreased and these could possibly deteriorate the soundness of forest areas by reducing the core areas which are habitats of species. In addition, there was intimate relations between ESV and forest landscape area. The results of this study can be detached standards for impartial judgements between the logic of development & conservation, and basic standards for the establishment of development plans, i. e. metropolitan-plans, which are adequately reflected ecosystem value.

• Journal of the Korean association of regional geographers
• /
• v.22 no.3
• /
• pp.615-627
• /
• 2016

The purpose of this study is to investigate the effect of urban sprawl and their ramifications, i.e. segmentation and hierarchization on the spatial structure as well as populational composition, focusing on Daecheon - Village and Daecheoncheon - Network in Busan, and to examine not only the solidarity between constituents transcending the segmented spaces but also the internal values operating inside through the Daecheoncheon - Network. Due to the large - scale housing development in the 1980s to 1990s, Daecheon - Village has been transformed from a rural village to a town. In this process, the original single space became segmented into Daecheon - Village and apartment complex. This spatial segmentation divided the populational composition into old natives and young immigrants. However, the Daecheoncheon - Network created by solidarity between the bodies of two localities enabled the residents to resolve the urgent issues of localities, recognizing their own space of living not as segmented and hierarchic but as the communal site of life and one village where they will live together. Daecheoncheon-Network was the movement and network to connect natives and immigrants transcending the segmented space and went so far as to make a motive to create one community with the value of 'symbiosis.'

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.28 no.4
• /
• pp.152-158
• /
• 2018

PVT (Physical Vapor Transport) method has advantages in producing high quality, large scale wafers where many researches are being carried out to commercialize nitride semiconductors. However, complex process variables cause various defects when it had non-equilibrium growth conditions. Annealing process after crystal growth has been widely used to enhance the crystallinity. It is important to set appropriate temperature, pressure, and annealing time to improve crystallinity effectively. In this study, the effect of the annealing conditions on the crystalline structure variation of the AlN single crystal grown by PVT method was investigated with synchrotron whitebeam X-ray topography, electron backscattered diffraction (EBSD), and Rietveld refinement. X-ray topography analysis showed secondary phases, sub-grains, impurities including carbon inclusion in the single crystal before annealing. EBSD analyses identified that sub-grains with slightly tilted basal plane appeared and the overall number of grains increased after the annealing process. Rietveld refinement showed that the stress caused by the temperature gradient during the annealing process between top and bottom in the hot zone not only causes distortion of grains but also changes the lattice constant.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2005.09a
• /
• pp.267-271
• /
• 2005

The goal of data mining is to extract new and useful knowledge from large scale datasets. As the amount of available data grows explosively, it became vitally important to develop faster data mining algorithms for various types of data. Recently, an interest in developing data mining algorithms that operate on graphs has been increased. Especially, mining frequent patterns from structured data such as graphs has been concerned by many research groups. A graph is a highly adaptable representation scheme that used in many domains including chemistry, bioinformatics and physics. For example, the chemical structure of a given substance can be modelled by an undirected labelled graph in which each node corresponds to an atom and each edge corresponds to a chemical bond between atoms. Internet can also be modelled as a directed graph in which each node corresponds to an web site and each edge corresponds to a hypertext link between web sites. Notably in bioinformatics area, various kinds of newly discovered data such as gene regulation networks or protein interaction networks could be modelled as graphs. There have been a number of attempts to find useful knowledge from these graph structured data. One of the most powerful analysis tool for graph structured data is frequent subgraph analysis. Recurring patterns in graph data can provide incomparable insights into that graph data. However, to find recurring subgraphs is extremely expensive in computational side. At the core of the problem, there are two computationally challenging problems. 1) Subgraph isomorphism and 2) Enumeration of subgraphs. Problems related to the former are subgraph isomorphism problem (Is graph A contains graph B?) and graph isomorphism problem(Are two graphs A and B the same or not?). Even these simplified versions of the subgraph mining problem are known to be NP-complete or Polymorphism-complete and no polynomial time algorithm has been existed so far. The later is also a difficult problem. We should generate all of 2$^n$ subgraphs if there is no constraint where n is the number of vertices of the input graph. In order to find frequent subgraphs from larger graph database, it is essential to give appropriate constraint to the subgraphs to find. Most of the current approaches are focus on the frequencies of a subgraph: the higher the frequency of a graph is, the more attentions should be given to that graph. Recently, several algorithms which use level by level approaches to find frequent subgraphs have been developed. Some of the recently emerging applications suggest that other constraints such as connectivity also could be useful in mining subgraphs : more strongly connected parts of a graph are more informative. If we restrict the set of subgraphs to mine to more strongly connected parts, its computational complexity could be decreased significantly. In this paper, we present an efficient algorithm to mine frequent subgraphs that are more strongly connected. Experimental study shows that the algorithm is scaling to larger graphs which have more than ten thousand vertices.