• 한국전자파학회논문지
• /
• 제20권11호
• /
• pp.1178-1185
• /
• 2009

본 논문은 2.5 MHz 간격으로 광대역 주파수를 발생하며 응답 시간이 400 nsec 이하인 초고속 이산(discrete) 주파수 합성기를 제안한다. 제안한 주파수 합성기는 고정 주파수 위상 제어 루프(PLL)와 주파수 분배기를 이용해 16개의 기준 신호를 생성하고, 이들을 선택하여 주파수 혼합하는 방식으로 710~1,610 MHz내에서 2.5 MHz 간격의 이산 주파수 신호를 고속으로 생성한다. 제작된 주파수 합성기의 주파수 천이 응답 시간은 평균 350 nsec, 고조파를 비롯한 모든 불요파 신호는 -60 dBc 이하, 위상 잡음 특성은 -94 dBc/Hz @100 Hz, 출력 세기는 평균 21.5 dBm, 평탄도는 2.65 dB 이하로 측정되었다. 또한 주파수 천이 응답 속도를 측정하는 새로운 기법의 측정 방법이 제안되었다.

• 한국주거학회논문집
• /
• 제20권4호
• /
• pp.31-38
• /
• 2009

In our country, the application of the design method of face to face dimension in wall-slab apartment has passed 10 years. So MC(Modular Coordination) design method has fixed in the design step to some degree. In Design and Work Execution of House the exclusive area of the apartment was calculated by face to face dimension. And the term of face to face dimension was known broadly to clients, construction company, etc. But design method of face to face dimension is not to simply extend the room size so far as wall depth in design process but to ensure the actual space should be considered with efficient use and assembly of building components. That is to say, Design method of face to face dimension is not to be defined by design step but to be related with construction and maintenance. However in construction process the point of face to face design method was not understood even now. So the purpose of this study was to extract the effect and question of face to face design method in construction process by survey of actual condition of structure and construction quality, and compare this result with existing construction method. The following project of this study is to survey of actual condition of interior components such as gypsum board, windows & doors etc. Therefore this study is to induce architectural long-life through architectural standardization construction and component's exchange, and, by the subject of the study on Apartment to have design guideline and basis for joining part between Gypsumboard and windows.

• 한국콘텐츠학회논문지
• /
• 제19권9호
• /
• pp.476-484
• /
• 2019

RFID 요킹증명은 한 쌍의 태그가 동시에 스캔 되었는지를 증명한다. 단일 리더에 의해 동시에 스캔 된 태그들은 서로 인접해 있기 때문에, 요킹증명은 태그가 부착된 객체들의 물리적 인접성을 검사해야 하는 응용에서 사용된다. 현재까지 제안된 대부분의 요킹증명 기법들은 인접한 태그에 관한 사전정보가 필수적이다. 만약 인접한 태그에 관한 정보를 수집하는 과정에서 오류가 발생하면, 이후의 증명들은 모두 검증에 실패하게 된다. 그러나 현재까지 인접한 태그에 관한 정보를 얻기 위한 구체적인 방법을 제시한 연구는 아직 없다. 이 연구에서 나는 요킹증명 기법을 위한 인접태그 정보 획득 기법인 TPIA을 제안한다. 제안기법은 스캐닝영역 결정과 스캐닝영역 검증의 두 단계로 구성된다. 스캐닝영역 결정 단계에서는 태그들의 위치와 전송범위를 고려하여 태그들을 스캔할 영역의 크기와 위치를 결정하며, 스캐닝영역 검증 단계에서는 고정된 위치의 참조태그를 사용하여 태그 스캐닝이 스캐닝영역 안에서 수행되었는지 여부를 검증한다. 나는 분석에서 제안하는 스캐닝영역이 정상적인 인접태그 정보의 획득을 보장하며, 스캐닝영역 검증이 스캐닝영역의 변형과 이탈을 감지하는 것을 보인다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.321-325
• /
• 2004

Retinopathy of Prematurity (ROP) is a common retinal neovascular disorder of premature infants. It can be characterized by inappropriate and disorganized vessel. This paper present a method for blood vessel detection on infant retinal images. The algorithm is designed to detect the retinal vessels. The proposed method applies a Lapalacian of Gaussian as a step-edge detector based on the second-order directional derivative to identify locations of the edge of vessels with zero crossings. The procedure allows parameters computation in a fixed number of operations independent of kernel size. This method is composed of four steps : grayscale conversion, edge detection based on LOG, noise removal by adaptive Wiener filter & median filter, and Otsu's global thresholding. The algorithm has been tested on twenty infant retinal images. In cooperation with the Digital Imaging Research Centre, Kingston University, London and Department of Opthalmology, Imperial College London who supplied all the images used in this project. The algorithm has done well to detect small thin vessels, which are of interest in clinical practice.

• 한국대기환경학회지
• /
• 제22권3호
• /
• pp.373-381
• /
• 2006

The catalytic combustion of toluene was investigated on the Cu-Mn oxide catalysts prepared by the deposition-precipitation method. Experiment of toluene combustion was performed with a fixed bed flow reactor in the temperature range of $100{\sim}280^{\circ}C$. Among the catalysts, 1.29Cu/Mn showed the most activity at $260^{\circ}C$. The deposition-precipitation method may be showed the potential to enhance the activity of catalysts. The catalysts were characterized by BET, scanning electron microscopy (SEM), temperature-programmed reduction (TPR), X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD) techniques. On the basis of catalyst characterization data, the results showed that the surface of catalysts by deposition-precipitation method had uniform distribution and smaller particle size, which enhanced the reduction capability of catalysts. The XRD results showed that $Cu_{1.5}Mn_{1.5}O_{4}$ spinel phase was made by deposition-precipitation method, and increased catalyst activity and redox characteristic. It was assumed that the reduction step of $Cu_{1.5}Mn_{1.5}O_{4}$ spinel phase progressed $Cu_{1.5}Mn_{1.5}O_{4}\;to\;CuMnO_{2},\;and\;Cu_{2}O\;to\;CuMn_{2}O_{4}\;and\;Cu$.

• Archives of Plastic Surgery
• /
• 제37권6호
• /
• pp.775-778
• /
• 2010

Purpose: Reconstruction of soft tissue defects with osteomyelitis in the lower third of the leg represents a challenge to plastic surgeons. Moreover, it is more arduous in multimorbid patients. One excellent option for reconstruction of these defects is to use a delayed distally based sural flap. Methods: We successfully used delayed distally based sural flap with a two-step procedure. During the first operation, radical debridement and elevation of flap were performed. The raised flap was fixed again at the donor site. The delay period ranged from seven to ten days. Between August 2008 and July 2009, we underwent operations for five patients using this technique. The size of flap varied from $10{\times}6\;cm$ to $12{\times}14\;cm$. Results: All flaps successfully survived. Partial skin loss of the grafted site was seen in two patients but no further surgical procedure was required for wound healing. Complaints of hypoesthesia on the lateral part of the foot was observed. In a three month follow-up period, hypoesthesia was resolved spontaneously. Conclusion: Delayed procedure improves the viability of distally based sural flap in high risk, critically multimorbid patients. We recommend that, if a two-stage operative approach is required, the delayed procedure should be considered.

• Journal of Information Processing Systems
• /
• 제13권5호
• /
• pp.1135-1148
• /
• 2017

Speech recognition is one of the fascinating fields in the area of Computer science. Accuracy of speech recognition system may reduce due to the presence of noise present in speech signal. Therefore noise removal is an essential step in Automatic Speech Recognition (ASR) system and this paper proposes a new technique called combined thresholding for noise removal. Feature extraction is process of converting acoustic signal into most valuable set of parameters. This paper also concentrates on improving Mel Frequency Cepstral Coefficients (MFCC) features by introducing Discrete Wavelet Packet Transform (DWPT) in the place of Discrete Fourier Transformation (DFT) block to provide an efficient signal analysis. The feature vector is varied in size, for choosing the correct length of feature vector Self Organizing Map (SOM) is used. As a single classifier does not provide enough accuracy, so this research proposes an Ensemble Support Vector Machine (ESVM) classifier where the fixed length feature vector from SOM is given as input, termed as ESVM_SOM. The experimental results showed that the proposed methods provide better results than the existing methods.

• 대한임베디드공학회논문지
• /
• 제17권3호
• /
• pp.129-138
• /
• 2022

Collecting and labeling sufficient training data, which is essential to deep learning-based visual inspection, is difficult for manufacturers to perform because it is very expensive. This paper presents a steel plate surface defect detection system with industrial-grade detection performance by training a small amount of steel plate surface images consisting of labeled and non-labeled data. To overcome the problem of lack of training data, we propose two data augmentation techniques: program-based augmentation, which generates defect images in a geometric way, and generative model-based augmentation, which learns the distribution of labeled data. We also propose a 4-step semi-supervised learning using pseudo labels and consistency training with fixed-size augmentation in order to utilize unlabeled data for training. The proposed technique obtained about 99% defect detection performance for four defect types by using 100 real images including labeled and unlabeled data.

• Asia pacific journal of information systems
• /
• 제29권4호
• /
• pp.771-788
• /
• 2019

Deep learning techniques such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) show superior performance in text classification than traditional approaches such as Support Vector Machines (SVMs) and Naïve Bayesian approaches. When using CNNs for text classification tasks, word embedding or character embedding is a step to transform words or characters to fixed size vectors before feeding them into convolutional layers. In this paper, we propose a parallel word-level and character-level embedding approach in CNNs for text classification. The proposed approach can capture word-level and character-level patterns concurrently in CNNs. To show the usefulness of proposed approach, we perform experiments with two English and three Korean text datasets. The experimental results show that character-level embedding works better in Korean and word-level embedding performs well in English. Also the experimental results reveal that the proposed approach provides better performance than traditional CNNs with word-level embedding or character-level embedding in both Korean and English documents. From more detail investigation, we find that the proposed approach tends to perform better when there is relatively small amount of data comparing to the traditional embedding approaches.

• 대한지하수환경학회지
• /
• 제6권1호
• /
• pp.23-32
• /
• 1999

Richards 방정식(RE) 해를 구하는 새로운 수치해석적인 방법으로 해적응격자(SAG)법을 개발하였다. SAG 법은 격자생성법을 이용하여 해의 구배가 큰 영역에 더 많은 수의 격자가 밀집되도록 일정한 수의 격자를 자동으로 재분배한다. 이 방법은 좌표변환기법을 이용하여 지배방정식인 RE를 새로운 좌표에서의 RE로 변환하고 유한차분법을 적용하여 방정식의 해를 구한다. 이때 격자점들의 이동은 변환된 RE에 수식으로 반영되기 때문에 고정된 격자점 을 갖는 변환된 영역에서는 해를 구하는 과정에서 내삽법이 불필요하게 된다. 따라서 SAG법은 불포화대에서의 지하수 침투과정을 모사할 때 습윤전선의 이동과 관련하여 발생하는 수치해석적 난제를 크게 개선할 수 있는 방법이다. SAG법과 고정격자를 이용하는 기존의 수정 Picard법을 비교하기 위하여 1차원 침투문제에 대한 수치실험을 실시하였다. 41개의 격자점을 이용한 SAG법은 201점의 고정격자법과 비교하였을 때, 해의 정확도에서는 비슷한 값을 보였으며 계산시간은 반으로 줄어들었다. SAG해의 질량평형과 수렴도는 시간간격 ($\Delta$t)과 해적응 격자생성에 사용된 가중모수 (${\gamma}$)에 크게 영향을 받는 것으로 나타났다. 따라서 SAG법을 이용하여 질량보존적이며 동시에 수렴하는 해를 구하기 위해서는 $\Delta$t와 ${\gamma}$를 자동으로 재조정하는 과정이 요구되며, 이러한 과정은 계산시간을 증가시키는 요인으로 작용할수도 있을 것이다. 본 연구에서 제시된 방법은 특별히 시간에 따른 전선의 이동을 다루는 불포화 유동 및 오염물 거동 문제에서 유용하게 사용될 수 있을 것으로 기대된다.