• Journal of the Korean Vacuum Society
• /
• v.17 no.3
• /
• pp.211-214
• /
• 2008

GaN-based light-emitting diodes (LEDs) of a 405 nm wavelength have been fabricated on a sapphire substrate by metal organic chemical vapor deposition (MOCVD). In order to reflect the photons, which are generated in the InGaN active region and emitted to the backside, to the front surface, a reflection layer was deposited onto the back of the substrate. Aluminum was used as the reflection layer and Al was deposited on the sample followed by Ti evaporation for firm adhesion of the reflection layer to the substrate. The light extraction efficiency was enhanced 52 % by adoption of the Ti-Al reflection layer.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.55-60
• /
• 2006

The proposed method uses FFT(Fast Fourier Transform) and neural networks in order to extract texts in real time. In general, text areas are found in the higher frequency domain, thus, can be characterized using FFT. The neural network are learned by character region(high frequency) and non character region(low frequency). The candidate text areas can be thus found by applying the higher frequency characteristics to neural network. Therefore, the final text area is extracted by verifying the candidate areas. Experimental results show a perfect candidate extraction rate and about 95% text extraction rate. The strength of the proposed algorithm is its simplicity, real-time processing by not processing the entire image.

• International Journal of Computer Science & Network Security
• /
• v.24 no.1
• /
• pp.119-126
• /
• 2024

Disease caused by the coronavirus (COVID-19) is sweeping the globe. There are numerous methods for identifying this disease using a chest imaging. Computerized Tomography (CT) chest scans are used in this study to detect COVID-19 disease using a pretrain Convolutional Neural Network (CNN) ResNet50. This model is based on image dataset taken from two hospitals and used to identify Covid-19 illnesses. The pre-train CNN (ResNet50) architecture was used for feature extraction, and then fully connected layers were used for classification, yielding 97%, 96%, 96%, 96% for accuracy, precision, recall, and F1-score, respectively. When combining the feature extraction techniques with the Back Propagation Neural Network (BPNN), it produced accuracy, precision, recall, and F1-scores of 92.5%, 83%, 92%, and 87.3%. In our suggested approach, we use a preprocessing phase to improve accuracy. The image was enhanced using the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm, which was followed by cropping the image before feature extraction with ResNet50. Finally, a fully connected layer was added for classification, with results of 99.1%, 98.7%, 99%, 98.8% in terms of accuracy, precision, recall, and F1-score.

• Proceedings of the Korean Radioactive Waste Society Conference
• /
• 2004.02a
• /
• pp.245-256
• /
• 2004

The extraction of three kinds of amido podands, N,N,N'N'-tetrabutyl-3-oxa-pentanedi- amide (TBDGA), N,N,N'N'-tetra-isobutyl-3-oxa-pentanediamide(TiBDGA) and N,N,N'N'-tetra- butyl-3,6-dioxa-oct-anediam- ide(TBDOODA) on U(VI),Pu(IV), Am(III), Eu(III) and other metal ions is studied in nitric acid solutions. 40%octanol-kerosene is chosen as diluents to eliminate third phase and emulsion. TBDGA and TiBDGA show extraction selectivity to An(III) and Ln(III) much higher than to U(VI) and Pu(IV). Fe, Ru and Mo is poorly extracted by the three kinds of amid podands in 2~3mol/L $HNO_3$ solutions. Aiming to eliminate interface crude when using simulated HLLW solution in the system of 0.2mol/L TBDGA/Octanol＋kerosene, acetohydroxyamic acid was adapted. Distribution ratio of zirconium was decreased when adding acetohydroxyamic acid in aqueous solution, and interface crude disappeared as mixing extractant with HLLW. The counter-current extraction test is carried out in a set of miniature mixer-settler, with 0.2mol/L TBDGA/ 40% octanol-kerosene as extractant to separate U(VI), Pu(IV), Am(III) and Eu(III) from simulated high level liquid waste(HLLW) solution. In battery A, lanthanides and actinides are coextracted into organic phase with the recovery of 99.98% for U(Ⅵ), >99.99% for Pu(IV), and >99.99% for Am(III) and Eu(III) respectively. In battery R1, 99.99% U, 86.2% Pu and a part of Am or Eu are stripped into aqueous phase by 0.2mol/L acetohydroxyamic acid (AHA) in 0.01mol/L $HNO_3$ solution. In battery $R_2$, Am, Eu and remained Pu are completely back-extracted by 0.2mol/L AHA. This separation process contains no salt reagent, and it is not necessary to dilute HLLW feed.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.18 no.10
• /
• pp.965-975
• /
• 2005

In the case of the fault in stator windings of a high voltage motor. it facilitates certain destructive characteristics in insulations. This will result in a decreased reliability in power supplies and will prevent the generation of electricity, which will result in huge economic losses. This study simulates motor windings using normal windings and four faulty windings for an actual fault in stator winding of a high voltage motor. The partial discharge signals produced in each faulty winding were measured using an 80 PF epoxy/mica coupler sensor. In order to quantified signal waves its a way of feature extraction for each faulty signal, the signal wave of winding was quantified to measure the degree of skewness shape and kurtosis, which are both types of statistical parameters, using a discrete wavelet transformation method for each faulty type. Wave types present different types lot each faulty type, and the skewness and kurtosis also present different quantified values. The result of feature extraction was used as a preprocessing stage to identify a certain fault in stater windings. It is evident that the type of faulty signals can be classified from the test results using faulty signals that were randomly selected from the signal, which was not applied in the training after the training and learning period, by applying it to a back-propagation algorithm due to the supervising and learning method in a neural network in order to classify the faulty type. This becomes an important basis for studying diagnosis methods using the classification of faulty signals with a feature extraction algorithm, which can diagnose the fault of stator windings in the future.

• Research in Plant Disease
• /
• v.10 no.1
• /
• pp.53-57
• /
• 2004

Cucumber green mottle mosaic virus (CGMMV) was a major pathogen of watermelon and had affected seriously to watermelon production in Korea. Rapid and sensitive detection method of CGMMV associated with bottle gourd (Lagenafia siceraria) seeds was developed by using RT-PCR in this study. A pair of primeri Wmfl and Wmrl, specific for CGMMV was designed from coat protein gene sequences of CGMMV-W and used for amplifying 420 bp product in RT-PCR. To simplify the virus extraction procedure and reduce an inhibitor from the extract for the RT-PCR, some methods using ethanol precipitation, double filtration, polyethylene glycol precipitation and phenol/chloroform/isoamyl alcohol extraction procedure were compared and the phenol/chloroform/isoamyl alcohol extraction procedure was selected by its enhanced sensitivity. This detection method using the selected extraction step and the primers for RT-PCR could reliably detect an infected level of one CGMMV-infested seed in 1,000 seeds. This rapid and sensitive RT-PCR assay provides auseful tool for the specific detection of CGMMV in bottle gourd seed samples containing high levels of back-ground inhibitors.

• Analytical Science and Technology
• /
• v.6 no.1
• /
• pp.131-139
• /
• 1993

The separation of gallium and indium from the matrix elements such as zinc and other ions, especially form Fe(III) ion was studied for the determination of trace level of them in zinc ores and zinc blendes by inductively coupled plasma atomic emission spectrometry(ICP-AES). Gallium and indium were extracted from the sample solution with a solvent of tributyl phosphate(TBP). The type and concentration of acid, interferences of other ions, the ratio of aqueous phase to organic phase, TBP concentration, sripping efficiency were optimized for the effective extraction. Gallium and indium were separated from other ions in the 5N hydrochloric acid solution of the samples by the extraction with 100% TBP. In this time, Fe(III) was reduced to Fe(II) with hydroxylamine hydrochloride to prevent its coextraction prior to the main extraxtion. After stripped from organic phase by the back-extraction with 0.02N HCl, they were determined in the aqueous phase by ICP-AES. This method was known to be quantitative from the overall extraction of more than 95%.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.32A no.8
• /
• pp.133-139
• /
• 1995

In this paper, the frequency separation locking and interval stabilization of 4-channel DFB-LDs have been demonstrated using a fiber Fabry-Perot filter with an free spectral range of 100GHz. Frequency fluctuation and locking range of each channel were appeared to be within 15MHz and over 12GHz, respectively. Back-reflection curve from the fiber Fabry-Perot filter was used for the extraction of an error signal in order to increase the number of accomodable channels and extinction ratio.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.625-627
• /
• 2002

본 논문에서는 의료영상 인식 기술 중 인식률이 뛰어나고 신뢰성 있는 대뇌출혈성 병변인식 시스템을 구현하기 위하여 신호처리 분야에서 주로 사용되는 Wavelet 변환과 신경망 기법을 이용하고자 한다. 그러나 신경망 기법에서 주로 사용되는 비선형 최적화기법인 Gradient descent BP는 최적화 문제점을 해결하기에는 수렴속도가 느리기 때문에 적합하지 않다. 따라서 본 논문에서는 기존 Gradient descent BP를 보완한 Levenberg-Marquardt Back-Propagation을 대뇌출혈성 병변인식에 적용하여 구현함으로써 총 50개의 패턴 중 45개의 영상이 인식에 성공하였고 전체 평균 인식률은 각각 90%와 87%의 인식률을 보였다.

• Proceedings of the Korean Institute of IIIuminating and Electrical Installation Engineers Conference
• /
• 2009.05a
• /
• pp.157-159
• /
• 2009

LED BLU(Back Light Unit), based on MCPCB(Metal Core Printed Circuit Board) with anodizing oxide dielectric layer and improved thermal dissipation property, are presented. Reflecting cups were also formed on the surface of the MCPCB such that optical coupling between neighboring chips were minimized for improving the photon extraction efficiency. LED chips were directly attached on the MCPCB by using the COB (Chip On Board) scheme.