• Journal of environmental and Sanitary engineering
• /
• v.14 no.3
• /
• pp.1-9
• /
• 1999

In the estimation of pollution load in water basin, a data information has generally used from surveyed data. A Geographic Information System(GIS) was adopted to evaluate the amount of pollution load in Anyang stream basin which is one of the major tributaries in the Han river flows through urban area. The digital maps of administrative boundary, stream network, sub-basin, soil type, and land-use for spatial data as well as attribute data were generated. And the database of sub-basins and pollution source was structured to estimate pollution load in Anyang stream basin by an Arc/Info GIS.As the results of this investigation, the pollution load of Mokgam-chun sub-basin was the highest amount. And that of Hagi-chun sub-basin and the fourth main stream sub-basin were also high amount in Anyang stream basin. In general, it was found that the pollution load generated from the upstream area in Kyunggi province was higher than that from downstream area in Seoul. Because the point and non-point source pollution load played very significant role in the deterioration of the water quality of the Anyang stream, an integrated approach to water quality management should be required for the sub-basins of high pollution load amount.

• MALSORI
• /
• v.55
• /
• pp.93-102
• /
• 2005

In this paper, sub-stream based eigenvoice method is proposed to overcome the weak points of conventional eigenvoice and dimensional eigenvoice. In the proposed method, sub-streams are automatically constructed by the statistical clustering analysis that uses the correlation information between dimensions. To obtain the reliable distance matrix from covariance matrix for dividing into optimal sub-streams, MAP adaptation technique is employed to the covariance matrix of training data and the sample covariance of adaptation data. According to our experiments, the proposed method shows $41\%$ error rate reduction when the number of adaptation data is 50.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.878-879
• /
• 2021

Most state-of-the-art CNNs for action recognition are based on a two-stream architecture: RGB frames stream represents the appearance and the optical flow stream interprets the motion of action. However, the cost of optical flow computation is very high and then it increases action recognition latency. We introduce a design strategy for action recognition inspired by a two-stream network and teacher-student architecture. There are two sub-networks in our neural networks, the optical flow sub-network as a teacher and the RGB frames sub-network as a student. In the training stage, we distill the feature from the teacher as a baseline to train student sub-network. In the test stage, we only use the student so that the latency reduces without computing optical flow. Our experiments show that its advantages over two-stream architecture in both speed and performance.

• Journal of the Korean earth science society
• /
• v.44 no.4
• /
• pp.291-306
• /
• 2023

Shallow groundwater in rural areas is primarily polluted by agricultural activities. Nitrate-nitrogen is an indicator of artificial pollution. In this study, the hydrochemical characteristics and nitrate-nitrogen pollution of shallow groundwater were examined in two agricultural villages (Hyogyo-ri and Sinan-ri) in Chungcheongnam-do Province, Korea. Physicochemical quality analysis of shallow groundwater and stream water in the field, and chemical analysis in the laboratory were conducted from July 2020 to October 2021. In Hygyo-ri and Sinan-ri villages, shallow groundwater mainly belonged to the Ca-Cl, Ca-H CO₃, Na-HCO₃, and Na-Cl types, whereas stream water predominantly belonged to the Ca-HCO₃ type. The nitrate-nitrogen concentration in shallow groundwater varied depending on the season, displaying an increased concentration of nitrate-nitrogen in the dry season compared to the rainy season. Stream water may be influenced by runoff into villages from the surrounding area, although both shallow groundwater and stream water are affected by artificial pollution. In addition, the nitrate-nitrogen concentration in stream water was lower than that in shallow groundwater.

• The Research Journal of the Costume Culture
• /
• v.12 no.3
• /
• pp.327-338
• /
• 2004

This study was attempted to check period of a fashion cycle and to suggest its main-stream and sub-stream by analyzing the fashion trends presented in Vogue Korea between 1997 and 2002. A content analysis method was used to analyze 24 volumes. The result showed that the fashion trend of 1997's was a turning point from the simplicity, which was a trend up to 1996's, to the romanticism. The fashion trends of 2002 seemed to be also a turning point from the new-romanticism which was added the hippism to the romanticism continued from 1997, to the sporticism. The result confirmed that the main-stream of fashion theme have been continued for 5 years as a regular cycle system and could be repeated. It also verified that some sub-stream which was affected by social, cultural and political situation led the fashion changes. The results of this study could be expected to reflect not only fashion design, merchandising and marketing strategy but also consumer purchase behavior of the future.

• The Journal of Engineering Geology
• /
• v.32 no.4
• /
• pp.559-569
• /
• 2022

This study investigated the acidification and mixing with seawater of groundwater, stream water, and reservoir water in the Hunghae area of Pohang City, as well as the source of water expelled to the stream by liquefaction induced by the Pohang earthquake on 15 November 2017. Geologically, the area consists of Tertiary sedimentary rocks. We collected six samples of groundwater, five of reservoir water, four of stream water, two of liquefaction water, and one of seawater to analyze the chemical composition and stable isotopes (𝛿D and 𝛿¹⁸O). Gogkang Stream flows eastward through the central part of the study area into the East Sea. The groundwater and reservoir water in the lower part of the stream are acidic (pH < 4), have a Ca(Mg)-SO₄ composition, and high concentrations of Al, Fe, and Mn, likely due to the oxidation of pyrite in Tertiary rocks. The groundwater in the upper part of the stream have a Ca(Na)-HCO₃(Cl) composition, indicating the mixing of seawater with the stream water. The 𝛿D and 𝛿¹⁸O isotope data indicate the isotopic enrichment of reservoir water by evaporation. Based on the chemical and isotopic data, it is inferred that the two samples of liquefaction water originated from alluvium water in a transition zone with stream water, and from deep and shallow groundwaters that has been infiltrated by seawater, respectively.

• Nuclear Engineering and Technology
• /
• v.51 no.8
• /
• pp.1871-1885
• /
• 2019

The deterministic MOC code STREAM of the Computational Reactor Physics and Experiment (CORE) laboratory of Ulsan National Institute of Science and Technology (UNIST), was initially designed for the calculation of pressurized water reactor two- and three-dimensional assemblies and cores. Since fast reactors play an important role in the generation-IV concept, it was decided that the code should be upgraded for the analysis of fast neutron spectrum reactors. This paper presents a coupled code - TULIP/STREAM, developed for the fast reactor assembly and core calculations. The TULIP code produces self-shielded multi-group cross-sections using a one-dimensional cylindrical model. The generated cross-section library is used in the STREAM code which solves eigenvalue problems for a two-dimensional assembly and a multi-assembly whole reactor core. Multiplication factors and steady-state power distributions were compared with the reference solutions obtained by the continuous energy Monte-Carlo code MCS. With the developed code, a sensitivity study of the number of energy groups, the order of anisotropic P_N scattering, and the multi-group cross-section generation model was performed on the k_eff and power distribution. The 2D core simulation calculations show that the TULIP/STREAM code gives a k_eff error smaller than 200 pcm and the root mean square errors of the pin-wise power distributions within 2%.

• Journal of Korean Society on Water Environment
• /
• v.37 no.4
• /
• pp.249-262
• /
• 2021

This study investigated the pollution characteristics of the main pollution zone in the Hwangguji watershed and the influence of the tributary on the main stream. The characteristics of the main pollution zone, including, the water quality index (WQI), stream rating, load duration curve (LDC), delivery load density (DLD), and contribution of the tributary to the mainstream, were evaluated by time-series visual heatmap. The WQI of the mainstream of Hwangguji was lowered to the poor (IV) level from the inflow point of Suwon stream (SW) and the LDC excess rate in the T-P was higher than that of BOD₅, especially for the wet season, suggesting that management of non-point source with T-P is preferred. The contribution (%) of the tributaries in the upstream section of Hwangguji watershed were BOD₅ 14.54%, TOC 15.67%, T-N 5.43%, and T-P 6.97%. In particular, the Suwon sewage treatment plant located in the mainstream showed a high contribution of BOD₅ (64.40%) and T-P (53.54%), respectively, due to the high discharge rate (6.019 m³/sec). Meanwhile, Sammi and Gal stream have a large impact on the mainstream with high DLD and poor WQI. Thus, both streams were considered as pollution hot spots. These results provide useful basic data for preparing more effective water quality improvement and management plans in the watershed.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.3
• /
• pp.99-104
• /
• 2022

Convolutional neural network (CNN) generally uses two-stream architecture RGB and optical flow stream for its action recognition function. RGB frames stream display appearance and optical flow stream interprets its action. However, the standard method of using optical flow is costly in its computational time and latency associated with increased action recognition. The purpose of the study was to evaluate a novel way to create a two sub-networks in neural networks. The optical flow sub-network was assigned as a teacher and the RGB frames as a student. In the training stage, the optical flow sub-network extracts features through the teacher sub-network and transmits the information to student sub-network for baseline training. In the test stage, only student sub-network was operational with decreased in latency without computing optical flow. Experimental results shows that our network fed only by RGB stream gets a competitive accuracy of 54.5% on HMDB51, which is 1.5 times better than that on R3D-18.

• Journal of Ecology and Environment
• /
• v.38 no.4
• /
• pp.505-515
• /
• 2015

The objectives of this study were to determine the effects of stream nutrient regime, N:P ratios and suspended solids on fish tolerance/trophic compositions and stream ecosystem health, based on multi-metric model, during 2008-2013. Also, stream ecosystem health was evaluated in relation to chlorophyll-a (CHL) as a measure of algal productivity or indicators of trophic state to water chemical parameters. Total number of sampled fish species were 50 and showed a decreasing trend from 2008 to 2013. The minnow of Zacco platypus, based on the catch per unit effort (CPUE), was the most dominant species (25.9%) among the all species. Spatial heterogeneity was evident in the fish tolerance guilds that showed the dominance of sensitive species (89%) in the headwaters (S1) and the dominance of tolerant species (57%) in the urban. These conditions were directly influenced by concentrations of nutrients and organic matter (COD). The N:P ratios, as a barometer of water pollution, had a negative linear function (R² = 0.40, P < 0.01) with CHL, and the ratios had an important role in changes of COD concentration (R² = 0.40, P < 0.01). Under the circumstances, the N:P ratio directly influenced the relative proportions of fish trophic/tolerance compositions. According to the regression analysis of omnivore (O_m) and insectivore sp. (I_n) on total nitrogen and total phosphorus, nitrogen had no significant influences (P > 0.05) to the two compositions, but phosphorus influenced directly the two guilds [slope (a) = -32.3, R² = 0.25, P < 0.01 in the I_n; a = 40.7, R² = 0.19, P < 0.01 in the O_m]. Such water chemistry and fish trophic guilds determined the stream ecosystem health, based on the multi-metric fish model.