• Journal of Korean Society on Water Environment
• /
• v.35 no.6
• /
• pp.589-597
• /
• 2019

The purpose of this study was improve accuracy the IOPs inversion model(IOPs-IM) developed in 2016 for phycocyanin(PC) concentration estimation in the Nakdong River. Additionally, two optimum models were developed and evaluated with 2017 measurement field spectral data for the Geum River and the Yeongsan River. The used measurement data for IOPs-IM analyzation was randomly classified as training and verification materials at the ratio of 2:1 in all data sets. Using the training data set from 2015-2017, accuracy results of the IOPs-IM generally improved for the Nakdong River. The RMSE(Root Mean Square Error) decreased by 14 % compared to 2016. For the GeumRiver, the results of the IOPs-IM were suitable, except for some point results in 2016. Results of the IOPs-IM in the Yeongsan River followed the overall 1:1 line and MAE(Mean Absolute Error) was lower than other rivers. But the RMSE and MAE values were higher. As a result of applying the validation data to the IOPs-IM, the accuracy of the Nakdong River was reduced to RMSE 17.7 % and MRE 16.4 %, respectively compared with 2016. However, the MRE(Mean Relative Error) was estimated to be higher by 400 % in the Geum River, and the RMSE was more than 100 mg/㎥ of the Yeongsan River. Therefore, it is necessary to get the continuously data with various sections of each river for obtain objective and reliable results and the models should be improved.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1131-1131
• /
• 2001

To develop a nondestructive quality evaluation technique of fruits, a K-mean algorism is applied to near infrared (NIR) spectroscopy of apples. The K-mean algorism is one of neural network partition methods and the goal is to partition the set of objects O into K disjoint clusters, where K is assumed to be known a priori. The algorism introduced by Macqueen draws an initial partition of the objects at random. It then computes the cluster centroids, assigns objects to the closest of them and iterates until a local minimum is obtained. The advantage of using neural network is that the spectra at the wavelengths having absorptions against chemical bonds including C-H and O-H types can be selected directly as input data. In conventional multiple regression approaches, the first wavelength is selected manually around the absorbance wavelengths as showing a high correlation coefficient between the NIR $2^{nd}$ derivative spectrum and Brix value with a single regression. After that, the second and following wavelengths are selected statistically as the calibration equation shows a high correlation. Therefore, the second and following wavelengths are selected not in a NIR spectroscopic way but in a statistical way. In this research, the spectra at the six wavelengths including 900, 904, 914, 990, 1000 and 1016nm are selected as input data for K-mean analysis. 904nm is selected because the wavelength shows the highest correlation coefficients and is regarded as the absorbance wavelength. The others are selected because they show relatively high correlation coefficients and are revealed as the absorbance wavelengths against the chemical structures by B. G. Osborne. The experiment was performed with two phases. In first phase, a reflectance was acquired using fiber optics. The reflectance was calculated by comparing near infrared energy reflected from a Teflon sphere as a standard reference, and the $2^{nd}$ derivative spectra were used for K-mean analysis. Samples are intact 67 apples which are called Fuji and cultivated in Aomori prefecture in Japan. In second phase, the Brix values were measured with a commercially available refractometer in order to estimate the result of K-mean approach. The result shows a partition of the spectral data sets of 67 samples into eight clusters, and the apples are classified into samples having high Brix value and low Brix value. Consequently, the K-mean analysis realized the classification of apples on the basis of the Brix values.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1032-1032
• /
• 2001

Artificial Neural Network (ANN) calibration techniques have been used commercially for agricultural applications since the mid-nineties. Global models, based on transmission data from 850 to 1050 nm, are used routinely to measure protein and moisture in wheat and barley and also moisture in triticale, rye, and oats. These models are currently used commercially in approx. 15 countries throughout the world. Results concerning earlier European ANN models are being published elsewhere. Some of the findings from that study will be discussed here. ANN models have also been developed for coarsely ground samples of compound feed and feed ingredients, again measured in transmission mode from 850 to 1050 nm. The performance of models for pig- and poultry feed will be discussed briefly. These models were developed from a very large data set (more than 20,000 records), and cover a very broad range of finished products. The prediction curves are linear over the entire range for protein, fat moisture, fibre, and starch (measured only on poultry feed), and accuracy is in line with the performance of smaller models based on Partial Least Squares (PLS). A simple bias adjustment is sufficient for calibration transfer across instruments. Recently, we have investigated the possible use of ANN for a different type of NIR spectrometer, based on reflectance data from 1100 to 2500 nm. In one study, based on data for protein, fat, and moisture measured on unground compound feed samples, dedicated ANN models for specific product classes (cattle feed, pig feed, broiler feed, and layers feed) gave moderately better Standard Errors of Prediction (SEP) compared to modified PLS (MPLS). However, if the four product classes were combined into one general calibration model, the performance of the ANN model deteriorated only slightly compared to the class-specific models, while the SEP values for the MPLS predictions doubled. Brix value in molasses is a measure of sugar content. Even with a huge dataset, PLS models were not sufficiently accurate for commercial use. In contrast an ANN model based on the same data improved the accuracy considerably and straightened out non-linearity in the prediction plot. The work of Mr. David Funk (GIPSA, U. S. Department of Agriculture) who has studied the influence of various types of spectral distortions on ANN- and PLS models, thereby providing comparative information on the robustness of these models towards instrument differences, will be discussed. This study was based on data from different classes of North American wheat measured in transmission from 850 to 1050 nm. The distortions studied included the effect of absorbance offset pathlength variation, presence of stray light bandwidth, and wavelength stretch and offset (either individually or combined). It was shown that a global ANN model was much less sensitive to most perturbations than class-specific GIPSA PLS calibrations. It is concluded that ANN models based on large data sets offer substantial advantages over PLS models with respect to accuracy, range of materials that can be handled by a single calibration, stability, transferability, and sensitivity to perturbations.

• Journal of Korean Society for Atmospheric Environment
• /
• v.17 no.5
• /
• pp.395-405
• /
• 2001

To make an assessment of the compatibility between DOAS and conventional point monitoring system (MCSAM-2: MS2), we investigated the concentrations of three criteria pollutants which include S $O_2$, N $O_2$, and $O_3$from a national monitoring station in Seoul during the periods of June 1999~August 2000. The average concentration values for the whole study period derived from hourly concentration data sets of those three species indicated that the mean differences between the two methods can be approximated as 18%. When the bias structure of two systems was evaluated through the computation of percent difference(PD) between the two such as ( $C_{DOAS}$- $C_{conventional}$ $C_{DOAS}$*100, differences between the two systems appeared to be quite systematic among different compounds. While the mode of bias peaked at 0~20% or 20~40% in terms of PD values, the cause of such positive bias mainly arised from generally enhanced concentration values of DOAS system. The structure of bias among different species was further assessed through linear regression analysis. Results of the analysis indicated that the dominant portions of differences observed from two monitoring systems can be accounted for by the systematic differences in their spanning and zeroing systems. S $O_2$(MS2)=0.6385 S $O_2$(DOAS)+2.0985($r^2$=0.7894) N $O_2$(MS2)=0.6548 N $O_2$(DOAS)+7.437($r^2$=0.7687) $O_3$(MS2)=1.0359 $O_3$(DOAS)-7.7885($r^2$=0.7944) The findings of slope values at around 0.64~0.65 from two species suggest that DOAS should respond more sensitively in upper bound concentration range. The offset values apart from zero indicate that more deliberate comparison needs to be made between these monitoring systems. However, based on the existence of strong correlations from at least 8,000 data points for each species of comparison, we were able to conclude that the compatibility of two monitoring systems is highly significant. With the improvement of calibration techniques for the DOAS system. its applicability for routine monitoring of airborne pollutant species is expected to be quite extendable.

• Journal of The Korean Astronomical Society
• /
• v.49 no.1
• /
• pp.25-35
• /
• 2016

Multiple color selection techniques are successful in identifying quasars from wide-field broadband imaging survey data. Among the quasars that have been discovered so far, however, there is a redshift gap at 5 ≲ z ≲ 5.7 due to the limitations of filter sets in previous studies. In this work, we present a new selection technique of high redshift quasars using a sequence of medium-band filters: nine filters with central wavelengths from 625 to 1025 nm and bandwidths of 50 nm. Photometry with these medium-bands traces the spectral energy distribution (SED) of a source, similar to spectroscopy with resolution R ~ 15. By conducting medium-band observations of high redshift quasars at 4.7 ≤ z ≤ 6.0 and brown dwarfs (the main contaminants in high redshift quasar selection) using the SED camera for QUasars in EArly uNiverse (SQUEAN) on the 2.1-m telescope at the McDonald Observatory, we show that these medium-band filters are superior to multi-color broad-band color section in separating high redshift quasars from brown dwarfs. In addition, we show that redshifts of high redshift quasars can be determined to an accuracy of Δz/(1 + z) = 0.002 - 0.026. The selection technique can be extended to z ~ 7, suggesting that the medium-band observation can be powerful in identifying quasars even at the re-ionization epoch.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.18 no.3
• /
• pp.76-88
• /
• 2015

This paper presents a classification method based on the combination of active learning with temporal contextual information extracted from past land-cover maps for the classification of crop cultivation areas. Iterative classification based on active learning is designed to extract reliable training data and cultivation rules from past land-cover maps are quantified as temporal contextual information to be used for not only assignment of training data but also relaxation of spectral ambiguity. To evaluate the applicability of the classification method proposed in this paper, a case study with MODIS time-series vegetation index data sets and past cropland data layers(CDLs) is carried out for the classification of corn and soybean in Illinois state, USA. Iterative classification based on active learning could reduce misclassification both between corn and soybean and between other crops and non crops. The combination of temporal contextual information also reduced the over-estimation results in major crops and led to the best classification accuracy. Thus, these case study results confirm that the proposed classification method can be effectively applied for crop cultivation areas where it is not easy to collect the sufficient number of reliable training data.

• Journal of the Korean Chemical Society
• /
• v.53 no.6
• /
• pp.717-726
• /
• 2009

Multivariate models were developed for the simultaneous spectrophotometric determination of copper (II), nickel (II) and zinc (II) in water with 1-(2-thiazolylazo)-2-naphthol as chromogenic reagent in the presence of Triton X-100. To overcome the drawback of spectral interferences, principal component regression (PCR) and partial least square (PLS) multivariate calibration approaches were applied. Performances were validated with several test sets, and their results were then compared. In general, no significant difference in analytical performance between PLS and PCR models. The root mean square error of prediction (RMSEP) using three components for $Cu^{2+}$, $Ni^{2+}$ and $Zn^{2+}$ were 0.018, 0.010, 0.011 ppm, respectively. Figures of merit such as sensitivity, analytical sensitivity, limit of detection (LOD) were also estimated. High reliability was achieved when the proposed procedure was applied to simultaneous determination of $Cu^{2+}$, $Ni^{2+}$ and $Zn^{2+}$ in synthetic mixture and tap water.

• Korean Journal of Agricultural Science
• /
• v.48 no.2
• /
• pp.299-310
• /
• 2021

Glycerol is a non-volatile compound with no aromatic properties that contributes significantly to the quality of wine by providing sweetness and richness of taste. In addition, it is also the third most significant byproduct of alcoholic fermentation in terms of quantity after ethanol and carbon dioxide. In this study, Fourier transform infrared (FT-IR) spectroscopy was employed as a fast non-destructive method in conjugation with multivariate regression analysis to build a model for the quantitative analysis of glycerol concentration in wine samples. The samples were prepared by using three varieties of red wine samples (i.e., Shiraz, Merlot, and Barbaresco) that were adulterated with glycerol in concentration ranges from 0.1 to 15% (v·v^-1), and subjected to analysis together with pure wine samples. A net analyte signal (NAS)-based methodology, called hybrid linear analysis in the literature (HLA/GO), was applied for predicting glycerol concentrations in the collected FT-IR spectral data. Calibration and validation sets were designed to evaluate the performance of the multivariate method. The obtained results exhibited a high coefficient of determination (R²) of 0.987 and a low root mean square error (RMSE) of 0.563% for the calibration set, and a R² of 0.984 and a RMSE of 0.626% for the validation set. Further, the model was validated in terms of sensitivity, selectivity, and limits of detection and quantification, and the results confirmed that this model can be used in most applications, as well as for quality assurance.

• Korean Journal of Remote Sensing
• /
• v.26 no.2
• /
• pp.251-262
• /
• 2010

In order to provide quantitative control of the standard products of Geostationary Ocean Color Imager (GOCI), on-board radiometric correction, atmospheric correction, and bio-optical algorithm are obtained continuously by comprehensive and consistent calibration and validation procedures. The calibration/validation for radiometric, atmospheric, and bio-optical data of GOCI uses temperature, salinity, ocean optics, fluorescence, and turbidity data sets from buoy and platform systems, and periodic oceanic environmental data. For calibration and validation of GOCI, we compared radiometric data between in-situ measurement and HyperSAS data installed in the Ieodo ocean research station, and between HyperSAS and SeaWiFS radiance. HyperSAS data were slightly different in in-situ radiance and irradiance, but they did not have spectral shift in absorption bands. Although all radiance bands measured between HyperSAS and SeaWiFS had an average 25% error, the 11% absolute error was relatively lower when atmospheric correction bands were omitted. This error is related to the SeaWiFS standard atmospheric correction process. We have to consider and improve this error rate for calibration and validation of GOCI. A reference target site around Dokdo Island was used for studying calibration and validation of GOCI. In-situ ocean- and bio-optical data were collected during August and October, 2009. Reflectance spectra around Dokdo Island showed optical characteristic of Case-1 Water. Absorption spectra of chlorophyll, suspended matter, and dissolved organic matter also showed their spectral characteristics. MODIS Aqua-derived chlorophyll-a concentration was well correlated with in-situ fluorometer value, which installed in Dokdo buoy. As we strive to solv the problems of radiometric, atmospheric, and bio-optical correction, it is important to be able to progress and improve the future quality of calibration and validation of GOCI.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.1
• /
• pp.15-33
• /
• 2021

In this study, the possibility of discriminating Fire blight (FB) infection tested using the hyperspectral imagery. The reflectance of healthy and infected leaves and branches was acquired with 5 nm of full width at high maximum (FWHM) and then it was standardized to 10 nm, 25 nm, 50 nm, and 80 nm of FWHM. The standardized samples were divided into training and test sets at ratios of 7:3, 5:5 and 3:7 to find the optimal bands of FWHM by the decision tree analysis. Classification accuracy was evaluated using overall accuracy (OA) and kappa coefficient (KC). The hyperspectral reflectance of infected leaves and branches was significantly lower than those of healthy green, red-edge (RE) and near infrared (NIR) regions. The bands selected for the first node were generally 750 and 800 nm; these were used to identify the infection of leaves and branches, respectively. The accuracy of the classifier was higher in the 7:3 ratio. Four bands with 50 nm of FWHM (450, 650, 750, and 950 nm) might be reasonable because the difference in the recalculated accuracy between 8 bands with 10 nm of FWHM (440, 580, 640, 660, 680, 710, 730, and 740 nm) and 4 bands was only 1.8% for OA and 4.1% for KC, respectively. Finally, adding two bands (550 nm and 800 nm with 25 nm of FWHM) in four bands with 50 nm of FWHM have been proposed to improve the usability of multispectral image sensors with performing various roles in agriculture as well as detecting FB with other combinations of spectral bands.