• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.686-688
• /
• 2020

본 논문에서는 U-Net 모델을 이용하여 정교하고 반복되는 패턴을 가진 인쇄물에 대한 비지도 학습을 통한 딥러닝 기반 이상치탐지(Anomaly Detection) 방법을 제안하였다. 인쇄물(카드)의 비정상 패턴 검출을 위하여 촬영한 영상으로부터 카드 영역을 분리한 이미지로 구성된 Dataset을 구축하였고 정상 이미지와 동일한 이미지를 출력하기 위해, 정상 이미지와 마스크 이미지 쌍의 Training dataset을 U-Net으로 학습하였다. Test dataset의 이미지를 입력으로 넣어 생성된 마스크 결과를 원본 마스크 이미지와 비교하여 이상 여부를 판단하는 본 논문의 방법이 정상, 비정상 인쇄물을 잘 구분하는 것을 확인하였다. 또한 정상과 비정상 이미지 각각을 학습한 지도학습 기반 CNN 분류 방법을 입력 영상과 복원 영상 간의 복원 오차를 비교하여 객체의 이상 여부를 판별하는 본 논문의 방법과 비교 평가하였다. 본 논문을 통해 U-Net을 사용하여 별도로 데이터에 대한 label 취득 없이 이상치를 검출할 수 있음을 확인할 수 있었다.

• Environmental and Resource Economics Review
• /
• v.17 no.4
• /
• pp.813-843
• /
• 2008

This paper investigates the presence of long-run and short-run price asymmetries in weekly gasoline prices from January 1997 to July 2008. In accordance with distribution channels, wholesale and retail stages are analyzed separately. An approach based on TAR and M-TAR cointegration tests, which entail matching asymmetric ECMs, is employed. For wholesale prices, asymmetries in the links with crude oil prices and exchange rates are found for both ECMs in the long-run and short-run. Exchange rates appear to play more significant role than crude oil prices in explaining the short-run price asymmetry. The rise in crude oil prices or exchange rates has statistically significant major impact on the increase of wholesale prices on the second week, not immediately as expected in the concept of 'rockets and feathers'. And asymmetrically, the fall does not have any statistically significant effect on the same period. The finding seems to be somewhat unusual. However, for retail prices, asymmetry m connection with wholesale prices is only revealed in the long-run. A symmetric price adjustment can be assumed in the short-run. Contrary to the long-run asymmetry found in the wholesale stage, in the retail stage, the speed of adjustment for negative deviations toward long-run equilibrium is faster than for positive ones, which is a phenomenon not favorable to consumers.

• Korean Journal of Remote Sensing
• /
• v.17 no.3
• /
• pp.231-242
• /
• 2001

In 2-pass differential SAR interferometry(DInSAR), the topographic phase signature can be removed by using a digital elevation model(DEM) to isolate the contribution of deformation from interferometric phase. This method has an advantage of no unwrapping process, but applicability is limited by precision of the DEM used. The residual phase in 2-pass differential interferogram accounts for error of DEM used in the processing provided that no actual deformation exits. The objective of this paper is a preliminary study to improve DEM precision using low precision DEM and 2-pass DInSAR technique, and we applied the 2-pass DInSAR technique to Asan area. ERS-1/2 tandem complex images and DTED level 0 DEM were used for DInSAR, and the precision of resulting DEM was estimated by a 1:25,000 digital map. The input DEM can be improved by simply adding the DInSAR output to the original low precision DEM. The absolute altitude error of the improved DEM is 9.7m, which is about the half to that of the original DTED level 0 data. And absolute altitude error of the improved DEM is better than that from InSAR technique, 15.8m. This approach has an advantage over the InSAR technique in efficiently reducing layover effects over steep slope region. This study demonstrates that 2-pass DInSAR can also be used to improve DEM precision.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.5 no.1
• /
• pp.19-27
• /
• 2007

Combined method of extraction chromatography and liquid scintillation analysis was established for determinating $^{55}Fe\;and\;^{63}Ni$ radioactivity in solid samples. Activated concrete wastes generated from dismantling KRR-2 were analysed. The sequential separation including precipitation and extraction chromatography resulted in the above 90% chemical recoveries of Fe. Above 62% recoveries of Ni were obtained by this procedure exception to 43.6 and 46.5% recoveries. The seperation and counting procedure was also confirmed with spiked samples of known quantity. The measured and spiked quantity were agreed with the 3.7% and 0.7% variations in the $^{55}Fe\;and\;^{63}Ni$ experiments, respectively. The radioactivities of $^{55}Fe$ in the dismantled concretes are shown from below MDA to maximum 362 Bq/g. The radioactivities of $^{63}Ni$ in all concrete samples are below MDA. The $^{63}Ni$ doesn't exist in dismantled concretes from KRR-2. The radioactivity of $^{55}Fe$ is decreased rapidly as the sampling depth is increased from the concrete surface.

• Journal of the Korean Chemical Society
• /
• v.27 no.4
• /
• pp.279-286
• /
• 1983

An apparatus was made for the electrodeposition of alpha emitting actinide nuclides, $^{207}Bi$ and $^{210}Po$. The electrodeposition was made on a polished stainless steel plate cathode. The anode was made of platinum wire and to stir the solution. With the ammonium chloride as electrolyte initial pH = 4, chloride concentration = 0.6M and solution volume = 15ml, a current of 1.5 ampere(current density = 0.59A/$cm^2$) was flowed for 100 minutes for the quantitative recovery of electrodeposition and on average recovery of 98.3% was obtained within ${\pm}$0.7% uncertainty. Alpha spectrometry of the electrodeposited sample showed alpha peaks from $^{210}Po, ^{234}U$ and $^{239}Pu$ having energy resolution (FWHM) of 18.3, 21.8 and 36.0 keV respectively. The electrodeposition and alpha spectrometry for a natural uranium sample of domestic origin gave $^{238}U : ^{234}U = 1 : 6.1{\times}10^{-5}$ and for a neutron-irradiated uranium sample did $^{238}U : ^{239}Pu : ^{241}Am = 100 : 0.0263 : 5.20{times}10^{-5}$. The result of $^{238}U$ determination in the irradiated sample by electrodeposition-alpha spectrometry was in accord within ${\pm}1.6%$ of relative error with the results of solid fluorimetry and mass spectrometry. For $^{239}Pu$ the result of electrodeposition-alpha spectrometry was in accord within ${\pm}$4.0% of relative error with the results of anion exchange separation and the thenoyltrifluoroacetone(TTA) extraction both followed by alpha spectrometries.

• Journal of Wetlands Research
• /
• v.16 no.2
• /
• pp.161-171
• /
• 2014

Recession of hydrograph gives a significant contribution to estimation of baseflow using rainfall-runoff models and baseflow separation methods, because recession affects baseflow. This study attempted to enhance the accuracy of streamflow predictions using a Soil and Water Assessment Tool (SWAT) model and to separate baseflow from the predicted streamflow. For this, this study used two scenarios: 1) to calibrate eleven parameters using an auto-calibration tool with the alpha factor obtained from RECESS (S1); and 2) to calibrate twelve SWAT parameters including alpha factor (one of SWAT parameters) using an auto-calibration tool (S2). Then, baseflow spearation from the predicted streamflow was conducted by using Web-based Hydrograph Analysis Tool (WHAT). The results show that there is no significant difference between Nash-Sutcliffe efficiency (NSE) values of S1 and S2 for calibrations to streamflow. However, calibrations to baseflow showed that NSEs are 0.777 for S1 and 0.844 for S2, which means a significant difference. Quantitatively compared to the observed streamflow, relative errors were 20.78 % for S1 and 6.59 % for S2. Finally, this study showed the importance of recession in baseflow separated from the predicted streamflow using a rainfall-runoff model.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.6
• /
• pp.5-14
• /
• 1994

This paper is about an experiment of speaker-independent automatic Korean spoken words recognition using Multi-Layered Perceptron and Error Back-propagation algorithm. The object words are 50 citynames of D.D.D local numbers. 43 of those are 2 syllables and the rest 7 are 3 syllables. The words were not segmented into syllables or phonemes, and some feature components extracted from the words in equal gap were applied to the neural network. That led independent result on the speech duration, and the PARCOR coefficients calculated from the frames using linear predictive analysis were employed as feature components. This paper tried to find out the optimum conditions through 4 differerent experiments which are comparison between total and pre-classified training, dependency of recognition rate on the number of frames and PAROCR order, recognition change due to the number of neurons in the hidden layer, and the comparison of the output pattern composition method of output neurons. As a result, the recognition rate of $89.6\%$ is obtaimed through the research.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.7
• /
• pp.529-536
• /
• 2020

Reaction Control System of the third stage of the Korean Space Launch Vehicle II conducts roll control and 3 axis control throughout third stage engine start, satellite separation, and collision and contamination avoidance maneuver. Reaction control system consumes its propellant in each thruster operation. Hence, loading of proper amount of the propellant is important for mission success. It is needed to have a rough estimation method of propellant consumption during the flight. In this paper, we developed a energy equation using pressure and temperature data which are acquired in the on-board reaction control system. We constructed a test system which is similar with the on-board reaction control system to verify the energy equation. Test results using deionized water were compared with estimated propellant consumption. We also conducted an error analysis of the energy equation. We also presented the propellant consumption result of a system level operation test.

• The Korean Journal of Applied Statistics
• /
• v.23 no.5
• /
• pp.923-932
• /
• 2010

One of the problems in designing the fifth Korean National Health and Nutrition Examination Survey(KNHNES) is the lack of an appropriate sampling frame. Due to the significant time difference, we expect eight severe sampling frame errors if we use the sampling frame obtained from the latest population and housing census that was conducted in 2005. Thus, the construction of an appropriate sampling frame for the fifth KNHNES is crucial for a successful survey. We considered the construction of a sampling frame that overcomes the limitations of the 2005 population and housing census based frame. For the construction of eight new sampling frames, we considered the use of multiple sampling frames in which the frame for the apartment households and the frame for the general households are obtained from different sources.

• Journal of Advanced Marine Engineering and Technology
• /
• v.39 no.9
• /
• pp.959-966
• /
• 2015

In this paper, we propose an indoor visible light communication(VLC) localization using Walsh code which can identify overlapped signals transmitted from the different LED sources as each of orthogonal signal at a receiver and using Manchester code which can eliminate the flicker of LED light and maintain a constant brightness. The proposed system can estimate the relative position of the receiver by using Lambertian radiation properties and trilateration method that are applied to the location information of fixed LED sources and the received signals from them. In order to verify the feasibility of the proposed system, we carried out the simulation in an indoor space with $6{\times}6{\times}1.5m^3$ installed LED lamps of 16. The simulation result shows that the proposed method achieves an average positioning error of 0.0536 m and a maximum positioning error of 0.2977 m.