• Journal of The Korean Society of Agricultural Engineers
• /
• v.58 no.2
• /
• pp.53-63
• /
• 2016

Monitoring of cyanobacteria bloom in reservoir systems is important for water managers responsible of water supply system. Cyanobacteria affect the taste and smell of water and pose considerable filtration problems at water use places. Harmful cyanobacteria bloom in reservoir have significant economic impacts. We develop a new method for estimating the cyanobacteria bloom using Landsat TM and ETM+ data. Developed model was calibrated and cross-validated with existing in situ measurements from Daecheong Reservoir's Water Quality Monitoring Program and Algae Alarm System. Measurements data of three stations taken from 2004 to 2012 were matched with radiometrically converted reflectance data from the Landsat TM and ETM+ sensor. Stepwise multiple linear regression was used to select wavelengths in the Landsat TM and ETM+ bands 1, 2 and 4 that were most significant for predicting cyanobacteria cell number and bio-volume. Based on statistical analysis, the linear models were that included visible band ratios slightly outperformed single band models. The final monitoring models captured the extents of cyanobacteria blooms throughout the 2004-2012 study period. The results serve as an added broad area monitoring tool for water resource managers and present new insight into the initiation and propagation of cyanobacteria blooms in Daecheong reservoir.

• Journal of Korea Multimedia Society
• /
• v.19 no.8
• /
• pp.1310-1319
• /
• 2016

Face detection is the first step in a wide range of face applications. However, detecting faces in the wild is still a challenging task due to the wide range of variations in pose, scale, and occlusions. Recently, many deep learning methods have been proposed for face detection. However, further improvements are required in the wild. Another important issue to be considered in the face detection is the computational complexity. Current state-of-the-art deep learning methods require a large number of patches to deal with varying scales and the arbitrary image sizes, which result in an increased computational complexity. To reduce the complexity while achieving better detection accuracy, we propose a fully convolutional network-based face detection that can take arbitrarily-sized input and produce feature maps (heat maps) corresponding to the input image size. To deal with the various face scales, a multi-scale network architecture that utilizes the facial components when learning the feature maps is proposed. On top of it, we design multi-task learning technique to improve detection performance. Extensive experiments have been conducted on the FDDB dataset. The experimental results show that the proposed method outperforms state-of-the-art methods with the accuracy of 82.33% at 517 false alarms, while improving computational efficiency significantly.

• Journal of the Korea Computer Graphics Society
• /
• v.21 no.3
• /
• pp.55-61
• /
• 2015

In this paper, we introduce a method to synthesize realistic make-up effects on input images efficiently. In particular, we focus on shading on the make-up effects due to the lighting and face curvature. By doing this, we can synthesize a wider range of effects realistically than the previous methods. To do this, the information about lighting information together with the normal vectors on all pixels over the face region in the input image. Since the previous methods that compute lighting information and normal vectors require relatively heavy computation cost, we introduce an approach to approximate lighting information using cascade pose regression process and normal vectors by transforming, rendering, and warping a standard 3D face model. The proposed method consumes much less computation time than the previous methods. In our experiment, we show the proposed approximation technique can produce naturally looking virtual make-up effects.

• Journal of the Korea Computer Graphics Society
• /
• v.23 no.4
• /
• pp.21-29
• /
• 2017

Previous visual analytics researches has focused on reducing the uncertainty of predicted results using a variety of interactive visual data exploration techniques. The main purpose of the interactive search technique is to reduce the quality difference of the predicted results according to the level of the decision maker by understanding the relationship between the variables and choosing the appropriate model to predict the unknown variables. However, it is difficult to create a predictive model which forecast time series data whose overall trends is unknown such as youth physical growth data. In this paper, we pro pose a novel predictive analysis technique to forecast the physical growth value in small pieces of time series data with un certain trends. This model estimates the distribution of data at a particular point in time. We also propose a visual analytics system that minimizes the possible uncertainties in predictive modeling process.

• Korean Journal of Agricultural Science
• /
• v.46 no.2
• /
• pp.257-271
• /
• 2019

The use of food grade hexane (FGH) for edible oil extraction is responsible for the presence of benzene in the crude oil. Benzene is a Group 1 carcinogen and could pose a serious threat to the health of consumer. However, its detection still depends on classical methods using chromatography which requires a rapid non-destructive detection method. Hence, the aim of this study was to investigate the feasibility of using Fourier transform infrared (FTIR) spectroscopy combined with multivariate analysis to detect and quantify the benzene residue in edible oil (sesame and cottonseed oil). Oil samples were adulterated with varying quantities of benzene, and their FTIR spectra were acquired with an attenuated total reflectance (ATR) method. Optimal variables for a partial least-squares regression (PLSR) model were selected using the variable importance in projection (VIP) and the selectivity ratio (SR) methods. The developed PLS models with whole variables and the VIP- and SR-selected variables were validated against an independent data set which resulted in $R^2$ values of 0.95, 0.96, and 0.95 and standard error of prediction (SEP) values of 38.5, 33.7, and 41.7 mg/L, respectively. The proposed technique of FTIR combined with multivariate analysis and variable selection methods can detect benzene residuals in edible oils with the advantages of being fast and simple and thus, can replace the conventional methods used for the same purpose.

• Economic Analysis
• /
• v.25 no.2
• /
• pp.132-176
• /
• 2019

We use a time-varying parameter vector auto regression (TVP-VAR) model to understand the impact of U.S. monetary policy normalization on Korean financial markets and capital accounts. The U.S. monetary policy is represented by the federal funds rate, term premium and credit spread. During the U.S. monetary contraction period of 2004 to 2006, changes in the federal funds rate presented negative pressure on Korean financial markets. The changes in federal funds rate also led to a simultaneous contraction in inward and outward capital flows. However, the effects of a federal funds rate shock has been reduced since 2015. On the other hand, the effects of U.S. term premiums is getting stronger after the period of quantitative easing (QE). The influence of the U.S. credit spread also significantly increased after the global financial crisis. Simulation results show that a rise in the U.S. credit spread, which can be triggered by a contractionary monetary policy, can pose a larger adverse impact on the Korean economy than a rise in the federal funds rate itself. As for capital flows, a U.S. monetary policy contraction causes an outflow of foreign investment, but the repatriation of overseas investment by Korean residents can offset this outflow.

• Journal of Korean society of Dental Hygiene
• /
• v.10 no.2
• /
• pp.255-268
• /
• 2010

Objectives : In the field of dental hygiene, infections pose a serious problem. This issue has left many patients and dental staff exposed to microbes with potentially far-reaching effects. The purpose of this study was to find solutions which could improve Dental health policies and promote improved methods for the control of infections. This survey was conducted between March 3, 2008 and March 30, 2008, in the metropolitan area. The research was carried out in dental clinics and dental hygienists were surveyed through a questionnaire. Methods : Collected data was examined using the SPSS 14.0 program, using frequency, mean and standard deviation analysis, T-test, one-way ANOVA, Scheffe's test and Duncan's test. Further analysis was given using Logistic Regression. Results : The Performance of Infection Control by dental hygienists in Dental Clinics was shown at $2.96{\pm}0.42$. Disinfection, sterilization and the wearing of masks is high at $3.65{\pm}0.64$, $3.64{\pm}0.65$. The perception of the importance of infection control is higher among dental hygienists than dentists. Of the dental hygienists surveyed, 83.9% were exposed to at least one accident while at work. Age and years of experience were important in relation to infection control practices. The rate of infections and the number of accidents experienced related to the dental hygienists performance of infection control. The performance of infection control is influenced by the dental hygienists own perceptions, and the perceptions of their dentists. The dental hygienist's working conditions and their beliefs related to infection control were shown to influence their performance of infection control practices. Conclusion : At the conclusion of this investigation, it was determined that systematic refresher training of infection control should be delivered through further education and various media. Dental health care workers should be encouraged to practice these action items presented in the training. For successful implementation of infection control in every dental health-care setting, it is highly demanded that effective safe-guard tools, strategic support and standardized action items against infection problems are developed.

• Journal of Environmental Health Sciences
• /
• v.49 no.5
• /
• pp.262-274
• /
• 2023

Background: Exposure levels for heavy metals such as lead (Pb), mercury (Hg), and cadmium (Cd) have increased due to human activities. They are known to be a public health concern. Objectives: This study aimed to determine the exposure levels to heavy metals in the blood and urine of South Korean adults and to present the contribution rate of exposure pathways using an exposure algorithm for men aged 19~64, women aged 19~64, and all seniors aged 65 or older. Methods: We analyzed data from the Korean National Environmental Health Survey (KoNEHS) Cycle 4 (2018~2020). A total of 2,646 participants aged ≥19 years were included. Multiple regression analysis was performed to determine the factors affecting heavy metal concentrations. The contribution rate was calculated by applying three exposure algorithms for ingestion, inhalation, and dermal exposure. Results: Factors that commonly affect heavy metal concentrations in blood and urine were gender and age. The main influencing factors for Pb and Cd were education level and smoking status, while frequency of fish consumption and of alcohol consumption were indicated to be the main influencing factors for mercury. The contribution rates of lead and cadmium from food ingestion were 78.03~79.62% and 88.39~92.89%, respectively. Additionally, the highest contribution for mercury was accounted for by food at 81.69~85.77%. As a result of the risk assessment, cadmium was found to pose a potential health risk a with total cancer risk (TCR) of more than 1×10^-6. Conclusions: The KoNEHS could be an important study for determining the level of exposure to heavy metals and their influencing factors. Integrated exposure to heavy metals could assess the main exposure pathways, and this methodology could be applied to exposure management of heavy metals.

• The korean journal of orthodontics
• /
• v.26 no.4
• /
• pp.414-420
• /
• 1996

Cephalometric measureements have disadvantage of representing cranio-facial structures in two dimension only and therefore they pose limitations in describing three-dimentional structures of cranio-facial region. More interests have been put on the correlation between the two planes. This study evaluated correlations between facial type score, which allows effects on malocclusion, growth change prediction and establishment of treatment method and prognosis, and measurements from submentovertex view. Cephalometric view and submentovertex view were taken of skeletal Class I adults with optimal profile and correlations between them have been observed. Following results were obtained: 1. To learn about factors that influence average condylar angulation, FACE, INT-CO-ANG, MN-CORPUS, CON-RATIO, GON-RATIO, MN-RATIO were used as variables and underwent multiple regression analysis. As a result, the following equation was obtained : CON-AVE=.l73(FACE)-.322(INT-CO-ANG)+36.34(GON-RATIO) +.420(MN-CORPUS) (($R^2=.85451$) 2. The following equation was obtained concerning facial type score. FACE= .050(CON-ANG)+.023(INT-CO-ANG)-.075(MN-CORPUS)($R^2=.31547$) 3. Among the submentovertex measurements, MN-CORPUS, CON-RATIO, GON-RATIO, MN-RATIO showed close correlations. (P<0.05) 4. Average condylar angualtions were $23.37^{\circ}$ on the right and $20.71^{\circ}$ on left. There was a difference between the two. FACE : facial type soore. CON-ANG: mean value of condylar angulation. CON-AVE: mean value of Rt. Lt condylar angulation. INT-CO-ANG : angle between Rt. Lt condylar axis. MN-CORPUS : angle formed between RT. Lt gonion & pogonion. CON-RATIO: lntercondylar distance/mandibular body length. GON-RATIO : intergonion distanoe/mandibular body length. MN-RATIO: lntermylohyoid distance/mandibular body length. MX-RATIO: intermaxillary tuberosity distance/ANS-PNS distance.