• Journal of Environmental Science International
• /
• v.24 no.6
• /
• pp.743-753
• /
• 2015

In this study, it is an object to develop a regression model for the estimation of TOC (total organic carbon) concentration using investigated data for three years from 2010 to 2012 in the Gam Stream unit watershed, and applied in 2009 to verify the applicability of the regression model. TOC and $COD_{Mn}$ (chemical oxygen demand) were appeared to be derived the highest correlation. TOC was significantly correlated with 5 variables including BOD (biological oxygen demand), discharge, SS (suspended solids), Chl-a (chlorophyll a) and TP (total phosphorus) of p<0.01. As a result of PCA (principal component analysis) and FA (factor analysis), COD, TOC, SS, discharge, BOD and TP have been classified as a first factor. TOCe concentration was estimated using the model developed as an independent variable $BOD_5$ and $COD_{Mn}$. R squared value between TOC and measurement TOC is 0.745 and 0.822, respectively. The independent variable were added step by step while removing lower importance variable. Based on the developed optimal model, R squared value between measurement value and estimation value for TOC was 0.852. It was found that multiple independent variables might be a better the estimation of TOC concentration using the regression model equation(in a given sites).

• FLOWER RESEARCH JOURNAL
• /
• v.19 no.4
• /
• pp.219-224
• /
• 2011

Phalaenopsis (P.) has various species, and some of them have strong fragrance. There are fragrant species such as P. bellina, P. violacea, P. schilleriana and used in breeding program for fragrant Phalaenopsis. This study was performed for establishment of fragrance analysis system using GC-MS and electronic nose in eight P. resources. We analyzed fragrant compound using the tissue of sepal, petal, column, and lip of P. '3010'. The percentage of the major compound was high in the petal and lip tissues. The main compound emitted from P. bellina was linalool (21.21%). It was possible that fragrance pattern could be analyzed among the resources using the electronic nose. Discriminant function analysis (DFA) was more useful than the principal component analysis (PCA) in statistics program. We utilized GC-MS method for the major compounds of flower from our breeding materials. This study would be useful to the fragrant analysis system for the fragrant orchid breeding in the future.

• Journal of the Korean Society of Food Culture
• /
• v.22 no.2
• /
• pp.246-253
• /
• 2007

Mineral and volatile flavor compounds of Pimpinella brochycarpa N., a perennial Korean medicinal plant of the Umbelliferae family, were analyzed by inductively coupled plasma-atomic emission spectroscopy (ICP-AES) and simultaneous steam distillation extract (SDE)-gas chromatography mass spectrometry (GC/MS), head space solid phase micro-extraction (HS-SPME)-GC/MS. Mineral contents of the stalks and leaves were compared and the flavor patterns of the fresh and the shady air-dried samples were obtained by the electronic nose (EN) with 6 metal oxide sensors. Principal component analysis (PCA) was carried out using the data obtained from EN. The 1st principal values of the fresh samples have + values and the shady air-dried have - values. The essential oil extracted from the fresh and the shady air-dried by SDE method contain 58 and 31 flavor compounds. When HS-SPME method with CAR/PDMS fiber and PDMS fiber were used, 34 and 21 flavor compounds. The principal volatile components of Pimpinella brachycarpa N. were ${\alpha}$-selinene, germacrene D, and myrcene.

• Journal of Environmental Science International
• /
• v.12 no.7
• /
• pp.735-744
• /
• 2003

The field observations on a seasonal characteristic of organic matter and its origin in the surface sediment were carried out at 35 stations in Deukryang bay, southern coast of Korean Peninsula from May 1995 to February 1996. The analytical parameters were mud temperature, ignition loss(IL), chemical oxygen demand(COD), pheopigment, sulfide and water content. The origin and seasonal dynamics of organic matter in Deukryang Bay were analyzed by COD/IL, COD/sulfide ratio and principal component analysis(PCA). As a results of the mud temperature fluctuated between 2.1$^{\circ}C$ with the lowest mean 4.6$^{\circ}C$ in winter and 27.6$^{\circ}C$ with the highest mean 25.5$^{\circ}C$ in summer. The range of ignition loss(IL) was from 3.1% in autumn to 21.5% in winter. Chemical oxygen demand(COD) showed the highest mean value of 8.45 mg/g dry in spring within the range of 2.90∼18.21 mg/g dry, while it showed the lowest value of 4.33 mg/g dry in autumn within the range of 0.67∼10.37 mg/g dry. Pheopigments showed the highest mean value of 9.04 $\mu\textrm{g}$/g dry in autumn within the range of 1.36∼20.44 $\mu\textrm{g}$/g dry, while it did the lowest mean value of 2.20 $\mu\textrm{g}$/g dry in summer within the range of 0.33∼11.36 $\mu\textrm{g}$/g dry. The range of total sulfide (H$_2$S) was from no detect(ND) to 3.30 mg/g dry in spring. And water content showed the annual mean value of 43.6% within the range of 23.6∼54.9%. The source of organic matter by COD/IL and COD/sulfide ratio in Deukryang Bay had been producted by primary producer in sea water areas except the areas effected by small stream, domestic and animal wastes. And the analytical results of PCA was able to be divided into three different regions. The former was characterized by the shallow depth and authigenic organic matter from phytoplankton in northwest area and northeastern inner bay, the secondary was done by deeper depth and allochthonous one from lands in southeast area and eastern entrance of bay, and the latter was done by authigenic one from the farm of seaweeds such as, sea cabbage, sea mustard etc in western entrance of bay. But a study on the relationship between sulfide and COD concentration in the northeastern inner bay which was characterized by the water stagnation will to take much more studying including major constituents of organic matter in the future.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.12
• /
• pp.1150-1158
• /
• 2010

In this study, the Polynomial-based Radial Basis Function Neural Networks is proposed as one of the recognition part of overall face recognition system that consists of two parts such as the preprocessing part and recognition part. The design methodology and procedure of the proposed pRBFNNs are presented to obtain the solution to high-dimensional pattern recognition problem. First, in preprocessing part, we use a CCD camera to obtain a picture frame in real-time. By using histogram equalization method, we can partially enhance the distorted image influenced by natural as well as artificial illumination. We use an AdaBoost algorithm proposed by Viola and Jones, which is exploited for the detection of facial image area between face and non-facial image area. As the feature extraction algorithm, PCA method is used. In this study, the PCA method, which is a feature extraction algorithm, is used to carry out the dimension reduction of facial image area formed by high-dimensional information. Secondly, we use pRBFNNs to identify the ID by recognizing unique pattern of each person. The proposed pRBFNNs architecture consists of three functional modules such as the condition part, the conclusion part, and the inference part as fuzzy rules formed in 'If-then' format. In the condition part of fuzzy rules, input space is partitioned with Fuzzy C-Means clustering. In the conclusion part of rules, the connection weight of pRBFNNs is represented as three kinds of polynomials such as constant, linear, and quadratic. Coefficients of connection weight identified with back-propagation using gradient descent method. The output of pRBFNNs model is obtained by fuzzy inference method in the inference part of fuzzy rules. The essential design parameters (including learning rate, momentum coefficient and fuzzification coefficient) of the networks are optimized by means of the Particle Swarm Optimization. The proposed pRBFNNs are applied to real-time face recognition system and then demonstrated from the viewpoint of output performance and recognition rate.

• Korean Journal of Environmental Biology
• /
• v.34 no.4
• /
• pp.292-303
• /
• 2016

This study described the spatial distributions of marine environmental factors such as water temperature, salinity, chlorophyll a concentration and turbidity, and characteristics of phytoplankton community such as species composition, standing crops and dominant species at 19 fishing ports around Jeju Island during the early summer of 2016. I analyzed bio-oceanographical characteristics using principal component analysis (PCA) of the environmental factors and biological parameters. Water temperature, salinity, chlorophyll a and turbidity ranged from 17.6 to $20.7^{\circ}C$, from 26.19 to 32.33 psu, from 0.76 to $7.13{\mu}g\;L^{-1}$, and from 0.51 to 14.49 FTU, respectively. A total of 51 species of phytoplankton belonging to 35 genera were identified. In particular, diatoms and dinoflagellates accounted for more than 56.8% and 27.4% of all the species, respectively. Moreover, the number of phytoplankton species was controlled by salinity. Phytoplankton cell density ranged from $2.9cells\;mL^{-1}$ to $185.9cells\;mL^{-1}$. The dominant species were Navicula spp. Stephanopyxis turris, Eutreptiella gymnastica and Mesodinium rubrum. Environmental factors and the phytoplankton community varied greatly between sampling sites. According to PCA, the biological oceanographic characteristics of the around Jeju Island were characterized by meteorological factors such as air temperature, precipitation and discharge of ground water during early summer.

• Resources Recycling
• /
• v.26 no.2
• /
• pp.46-55
• /
• 2017

In this study, the design methodology of Radial Basis Function Neural Networks is developed with the aid of Laser Induced Breakdown Spectroscopy and also applied to the practical plastics sorting system. To identify black plastics such as ABS, PP, and PS, RBFNNs classifier as a kind of intelligent algorithms is designed. The dimensionality of the obtained input variables are reduced by using PCA and divided into several groups by using K-means clustering which is a kind of clustering techniques. The entire data is split into training data and test data according to the ratio of 4:1. The 5-fold cross validation method is used to evaluate the performance as well as reliability of the proposed classifier. In case of input variables and clusters equal to 5 respectively, the classification performance of the proposed classifier is obtained as 96.78%. Also, the proposed classifier showed superiority in the viewpoint of classification performance where compared to other classifiers.

• Korean Journal of Environmental Biology
• /
• v.32 no.1
• /
• pp.75-87
• /
• 2014

This study describes about the spatio-temporal distributions in phytoplankton community such as species composition, standing crop and dominant species from May 2006 to February 2007 in the Jangsu bay and the northwestern parts of Gamak bay. Based on the principal component analysis (PCA) of the environmental factors as well as biological parameters, the bio-oceanographical characteristics were analysed. A total of 83 species of phytoplankton belonging to 47 genera were identified. Whereas diatoms and dinoflagellates occupied more than 65% and 30% of total species, respectively. The annual dominant species were Chaetoceros affinis, Paralia sulcata and Bacillaria paxillifera in spring, Chaetoceros didymus, Ch. affinis and Octactis octonaria in summer, Skeletonema costatum-like species and B. paxillifera in autumn. Moreover phytoplankton cell density was ranged between 3.1 $cells{\cdot}mL^{-1}$ in spring and 521.0 $cells{\cdot}mL^{-1}$ in winter. It fluctuated with an annual mean of 76.0 $cells{\cdot}mL^{-1}$ between the lowest value of 7.6 $cells{\cdot}mL^{-1}$ in spring and the highest value of 220.2 $cells{\cdot}mL^{-1}$ by Skeletonema costatum-like species in winter. Briefly, the phytoplankton cell density in the mixing seasons was higher in comparison with the other seasons. According to the PCA, the biological oceanographic characteristics of the Jangsu bay was affected by the introduction of outside seawater particularly in temperature increasing seasons, and the other seasons, it may be described the light intensity, and mix between inner and outer bay sea waters.

• International Journal of Control, Automation, and Systems
• /
• v.3 no.4
• /
• pp.571-579
• /
• 2005

In this paper, we proposed an algorithm for arrhythmia classification, which is associated with the reduction of feature dimensions by linear discriminant analysis (LDA) and a support vector machine (SVM) based classifier. Seventeen original input features were extracted from preprocessed signals by wavelet transform, and attempts were then made to reduce these to 4 features, the linear combination of original features, by LDA. The performance of the SVM classifier with reduced features by LDA showed higher than with that by principal component analysis (PCA) and even with original features. For a cross-validation procedure, this SVM classifier was compared with Multilayer Perceptrons (MLP) and Fuzzy Inference System (FIS) classifiers. When all classifiers used the same reduced features, the overall performance of the SVM classifier was comprehensively superior to all others. Especially, the accuracy of discrimination of normal sinus rhythm (NSR), arterial premature contraction (APC), supraventricular tachycardia (SVT), premature ventricular contraction (PVC), ventricular tachycardia (VT) and ventricular fibrillation (VF) were $99.307\%,\;99.274\%,\;99.854\%,\;98.344\%,\;99.441\%\;and\;99.883\%$, respectively. And, even with smaller learning data, the SVM classifier offered better performance than the MLP classifier.

• Korean Journal of Agricultural Science
• /
• v.45 no.3
• /
• pp.455-461
• /
• 2018

Deoxynivalenol (DON), a Fusarium mycotoxin, causes health hazards for both humans and livestock. Therefore, the aim of this study was to investigate the metabolic profiles of the liver, serum, and urine of piglets fed DON using proton nuclear magnetic resonance ($^1H-NMR$) spectroscopy. The $^1H-NMR$ spectra of the liver, serum, and urine samples of the piglets provided with feed containing 8 mg DON/kg for 4 weeks were aligned and identified using the icoshift algorithm of MATLAB $R^2013b$. The data were analyzed by multivariate analysis and by MetaboAnalyst 4.0. The DON-treated groups exhibited discriminating metabolites in the three different sample types. Metabolic profiling by $^1H-NMR$ spectroscopy revealed potential metabolites including lactate, glucose, taurine, alanine, glycine, glutamate, creatine, and glutamine upon mycotoxin exposure (variable importance in the projection, VIP > 1). Forty-six metabolites selected from the principal component analysis (PCA) helped to predict sixty-five pathways in the DON-treated piglets using metabolite sets containing at least two compounds. The DON treatment catalyzed the citrate synthase reactions which led to an increase in the acetate and a decrease in the glucose concentrations. Therefore, our findings suggest that glyceraldehyde-3-phosphate dehydrogenase, citrate synthase, ATP synthase, and pyruvate carboxylase should be considered important in piglets fed DON contaminated feed. Metabolomics analysis could be a powerful method for the discovery of novel indicators underlying mycotoxin treatments.