• Journal of IKEEE
• /
• v.17 no.4
• /
• pp.459-468
• /
• 2013

Under nonlinear characteristics of frames, we propose the frame interpolation using GRNN to enhance the visual picture quality. By full search with block size of 128x128~1x1 to reduce blocky artifact and image overlay, we select the frame having block of minimum error and re-estimate the nonlinear moving vector using GRNN. We compare our scheme with forward(backward) motion compensation, bidirectional motion compensation when the object movement is large or the object image includes zoom-in and zoom-out or camera focus has changed. Experimental results show that the proposed method provides better performance in subjective image quality compared to conventional MCFI methods.

• Journal of the Korea Society for Simulation
• /
• v.23 no.3
• /
• pp.19-25
• /
• 2014

Demand forecasting is the basis of management activities including marketing strategy. Especially, the demand of a part is remarkably important in supply chain management (SCM). In the fields of various industries, the part demand usually has the intermittent characteristic. The intermittent characteristic implies a phenomenon that there frequently occurs zero demands. In the intermittent demands, non-zero demands have large variance and their appearances also have stochastic nature. Accordingly, in the intermittent demand forecasting, it is inappropriate to apply the traditional time series models and/or cause-effect methods such as linear regression; they cannot describe the behaviors of intermittent demand. Markov bootstrap method was developed to forecast the intermittent demand. It assumes that first-order autocorrelation and independence of lead time demands. To release the assumption of independent lead time demands, this paper proposes a modified bootstrap method. The method produces the pseudo data having the characteristics of historical data approximately. A numerical example for real data will be provided as a case study.

• The Journal of the Korean dental association
• /
• v.54 no.12
• /
• pp.971-984
• /
• 2016

The increase rate of dentists' competition is very fast at metropolitan areas in South Korea. We compare metropolitan and rural parameters to investigate the relation between competition and revenue variation. The competition and revenue variables of 73 metropolitan and 75 rural areas were calculated from 2010 Census of Service Industry microdata which include non-insurance revenues of dental clinics. Independent sample t-test results showed that the level of competition among dental clinics in metropolitan areas is higher. The lowest and the low ranked revenues are higher in rural areas. The highest and the average revenues are higher in metropolitan areas. But, 25 percentile and median revenues has no significant difference between two areas. Simple log linear regression results showed that the number of clinics could explain the distribution of revenues in both areas better than the density of active dentists and Herfindahl-Hirschman index. In the areas with many clinics have high maximum and average revenues and low minimum revenues. The increasing rate of maximum revenues is higher in metropolitan areas though the decreasing rate of minimum revenues is higher in rural areas. Metropolitan areas have higher Gini coefficients than rural areas, but the increasing rate of Gini coefficients is lower than rural areas. Findings from this study are useful reference when the dentists select the opening areas. One is that the median revenues between metropolitan and rural areas have no significant difference. The other is that the rural areas ensure the more stable and uniform revenues. The results would help to relieve the consumptive competition among dentists and to achieve the distributional efficiency of dental human resources.

• Asian-Australasian Journal of Animal Sciences
• /
• v.31 no.9
• /
• pp.1431-1441
• /
• 2018

Objective: This study was conducted to assess the effects of replacing rice straw with different proportions of cassava pulp on growth performance, feed intake, digestibility, rumen microbial population, energy partitioning and efficiency of metabolizable energy utilization in beef cattle. Methods: Eighteen yearling Thai native beef cattle (Bos indicus) with an average initial body weight (BW) of $98.3{\pm}12.8kg$ were allocated to one of three dietary treatments and fed ad libitum for 149 days in a randomized complete block design. Three dietary treatments using different proportions of cassava pulp (100, 300, and 500 g/kg dry matter basis) instead of rice straw as a base in a fermented total mixed ration were applied. Animals were placed in a metabolic pen equipped with a ventilated head box respiration system to determine total digestibility and energy balance. Results: The average daily weight gain, digestible intake and apparent digestibility of dry matter, organic matter and non-fiber carbohydrate, total protozoa, energy intake, energy retention and energy efficiency increased linearly (p<0.05) with an increasing proportion of cassava pulp in the diet, whereas the three main types of fibrolytic bacteria and energy excretion in the urine (p<0.05) decreased. The metabolizable energy requirement for the maintenance of yearling Thai native cattle, determined by a linear regression analysis, was $399kJ/kg\;BW^{0.75}$, with an efficiency of metabolizable energy utilization for growth of 0.86. Conclusion: Our results demonstrated that increasing the proportion of cassava pulp up to 500 g/kg of dry matter as a base in a fermented total mixed ration is an effective strategy for improving productivity in zebu cattle.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.2
• /
• pp.40-48
• /
• 1996

In this paper, a 40 dimensional segment vector with 4 frame and 7 frame width in every monosyllable interval was compressed into a 10, 14, 20 dimensional vector using K-L expansion and neural networks, and these was used to speech recognition feature parameter for CHMM. And we also compared them with CHMM added as feature parameter to the discrete duration time, the regression coefficients and the mixture distribution. In recognition test at 100 monosyllable, recognition rates of CHMM +${\bigtriangleup}$MCEP, CHMM +MIX and CHMM +DD respectively improve 1.4%, 2.36% and 2.78% over 85.19% of CHMM. And those using vector compressed by K-L expansion are less than MCEP + ${\bigtriangleup}$MCEP but those using K-L + MCEP, K-L + ${\bigtriangleup}$MCEP are almost same. Neural networks reflect more the speech dynamic variety than K-L expansion because they use the sigmoid function for the non-linear transform. Recognition rates using vector compressed by neural networks are higher than those using of K-L expansion and other methods.

• Environmental Engineering Research
• /
• v.11 no.1
• /
• pp.45-53
• /
• 2006

Conventional coagulation is still the main treatment process for algae removal in water treatment. The coagulation efficiency can be significantly improved by the preoxidation of algae-containing water. Jar test was conducted to determine the optimal condition for the removal of diatoms, especially Cyclotella sp. by preoxidation and the subsequent coagulation. The effects of various concentration of PAC (Polyaluminum chloride) on coagulation with and without preoxidation using chlorine or potassium permanganate at different pHs (7.7 and 9.0) were evaluated. At pH 7.7, preoxidation with 2ppm $Cl_2$ followed by coagulation with 7.5 ppm PAC coagulant could reduce Cyclotella sp. concentration by 86%. At pH 9.0, preoxidation with 1 mg $KMnO_4/L$ followed by coagulation with 12.5 ppm PAC coagulant reduced Cyclotella sp. concentration by 85%. Non-linear regression was applied to determine the optimal condition. At pH 7.7 and 9.0, R was over 0.9, respectively. The pH of algal blooming water is over 9.0. Algae (diatom; Cyelotella sp.) can be controlled in the following ways: preoxidation with 1 mg $KMnO_4/L$ followed by coagulation with 12.5 ppm PAC coagulant can remove 80% algae from water. If water pH is adjusted to 7.7, it was expected that less amount of coagulant (7.5 or 10 mg PAC /L) after preoxidation ($Cl_2$ 2 ppm or $KMnO_4$ 0.33, 1 ppm) would be needed to achieve similar level of algae removal. The oxidation with 0.33ppm $KMnO_4$ followed by coagulation with 7.5 ppm PAC coagulant was preferable due to cost-effectiveness of treatment condition and color problem after treatment.

• Journal of Korean Society on Water Environment
• /
• v.22 no.5
• /
• pp.768-783
• /
• 2006

Biological stream environment assessment has been an essential trend of aquatic ecosystems among advanced countries, because the chemical water quality assessment has limitations that only represent temporal water environmental status. Benthic macroinvertebrates are the most popular bioindicator group useful to biological water quality assessment for a long time. In addition, a number of indices using benthic macroinvertebrates have recently been developed not only for water quality assessment but also for ecological health assessment. In this paper Korean Saprobic Index (KSI) based on Zelinka-Marvan's saprobic valency concept is developed by applying DIN 38410 in Germany. The KSI value was widely applicable to most Korean streams with the exception of sand-based large streams, for example main stream of Nakdong River. But as a result of non-linear regression analysis the correlation between KSI and $BOD_5$ concentration was high and KSI reflected water quality alteration for mid and long-term period. In spite of applicability of KSI, it should be supplemented and developed to make it strong by accumulating field data as well as to manage integrated water environment in general through continuous biomonitoring.

• Molecular & Cellular Toxicology
• /
• v.4 no.4
• /
• pp.331-337
• /
• 2008

This study was conducted to quantitatively evaluate the recovery effects of herbal medicines on acetaminophen-induced hepatotoxicity. In the present study, the recovery effects of 251 herbal medicines on THLE-2 cells that had been damaged by acetaminophen were evaluated using an MTS assay. THLE-2 cells were cultured in 96-well plates and then pretreated with or without 60 ${\mu}M$ acetaminophen (${IC}_{50}$ value: 35.84) for 1 hr. Next, different herbal medicines were added to the wells, after which the cells were reincubated at $37^{\circ}C$ for 24 hr. After first round of screening, the candidate herbal medicines were selected based on a recovery rate of greater than 40% and their efficacy were then determined by dose response kinetic analysis. Among these extracts, 8 herbal medicines (Terminalia chebula, Pueraria lobata, Acronychia laurifolia, Lopatherum gracile, Oroxylum indicum, Cynanchum atratum, Senecio scandens, and Sophora flavescens) had a strong recovery effect on acetaminophen-induced damage in THLE-2 cells. Dose response non-linear regression analysis demonstrated that Senecio scandens showed the best recovery rate (98%), and that its ${EC}_{50}$ was 19.54 ng/mL. Additional studies of these herbal medicines should be conducted to determine if they possess novel therapeutic agents for the prevention or treatment of liver disorders.

• Health Policy and Management
• /
• v.22 no.3
• /
• pp.451-465
• /
• 2012

In order to strengthen assurance of National Health Insurance, co-payment should be reduced. This can happen with collaborative efforts of patients, medical institutes, and government altogether at the same time. This research applied Dutton(1986)'s medical service research model with high R-square, and analyzed 2008 Korea Health Panel Data (Beta Version 1), that was examined by Korea Institute for Health and Social Affairs and National Health Insurance, in order to figure out influential variables on co-payment. In result of Multiple Linear Regression Analysis, R-square was 46.7%, the older the age, the patients who had surgery, the longer days of hospital treatment are, the higher gross income of a household is, the more hospitalized in upper grade general hospitals, and the more upper grade rooms and selecting a doctor are used. The results have statistical significance. When conducting research applying medical service research model, there is a need to apply Dutton(1986)'s medical service research model with high R-square. In order to strengthen assurance of National Health Insurance, first conditions should be that patients are hospitalized in upper grade general hospital, and at the same time, are patients who had surgery with long stay of hospitalization. In addition, if proven that patients used upper grade rooms and selecting a doctor due to lack of regular treatment and rooms, for certain number of days of such hospitalization, it is suggested to be provided with health care insurance in upper grade rooms and selecting a doctor in calculating co-payment limit.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.38 no.6
• /
• pp.347-352
• /
• 2005

These experiments investigated the conditioning pattern and the nitrification efficiency of a fluidized sand biofilter (FSB) for seawater application. The FSB fed artificial nutrient was fully conditioned within 22 weeks. The maximum nitrification efficiency of the FSB was achieved at a superficial water velocity (SWV) of 1.0 cm/sec. After fixing the superficial water velocity at 1.0 cm/sec, the nitrification rates of the FSB were assessed at 3 total ammonia nitrogen (TAN) loading rates (250, 500, 1,000 g TAN/$m^3$/day) and 3 water temperatures (12, 16, $20^{\circ}C$). The TAN concentration in the simulated culture tank ranged from 2.87 to 9.72 mg/L at TAN loading rate of 1,000 g TAN/$m^3$/day, while that ranged from 0.45 to 1.26 mg/L at TAN loading rate of 500 g TAN/$m^3$/day. The ranges of TAN concentration in the former were too high for aquatic organisms and those in the latter were acceptable. Therefore, the safe TAN loading rate for the FSB in seawater conditions was decided as 500 g TA/$m^3$/day. From these results, daily TAN removal rates (g TAN/$m^3$/day) of FSB under conditions of inlet TAN concentration (C, mg/L) and water temperature (T, $^{\circ}C$) were calculated by the following non-linear multi-regression equation: TAN removal rate: f(z)=-1,311.295+655.714LnT+225.775LnC ($r^2=0.962$).