• 센서학회지
• /
• 제5권2호
• /
• pp.55-60
• /
• 1996

The growth and microwave plasma assisted nitrogen doping of ZnSe by low pressure organometallic vapor phase epitaxy(OMVPE) has been investigated in a vertical downflow reactor equipped with a laser interferometer for in-situ growth rate measurements. Particular emphasis is placed on understanding growth characteristics of $H_{2}Se$ and the new adduct source dimethylzinc:triethyllamine($DMZn:NEt_{3}$) as compared with those obtained with $H_{2}Se$ and DMZn. At lower temperatures ($<300^{\circ}C$) and pressures(<30Torr), growth rates are higher with the adduct source and the surface morphology is improved relative to films synthesized with DMZn. Hall measurements and photoluminescence spectra of the grown films demonstrate that DMZn and $DMZn:NEt_{3}$ produce material with comparable electronic and optical properties. Microwave plasma decomposition of ammonia is investigated as a possible approach to increasing nitrogen incorporation in ZnSe and photoluminescence spectra are compared to those realized with conventional ammonia doping.

• Industrial Engineering and Management Systems
• /
• 제13권4호
• /
• pp.421-431
• /
• 2014

The market basket data in the form of a binary user-item matrix or a binary item-user matrix can be modelled as a binary classification problem. The binary logistic regression approach tackles the binary classification problem, where principal components are predictor variables. If users or items are sparse in the training data, the binary classification problem can be considered as a cold-start problem. The binary logistic regression approach may not function appropriately if the principal components are inefficient for the cold-start problem. Assuming that the market basket data can also be considered as a special regression problem whose response is either 0 or 1, we propose three supervised learning approaches: random forest regression, random forest classification, and elastic net to tackle the cold-start problem, comparing the performance in a variety of experimental settings. The experimental results show that the proposed supervised learning approaches outperform the conventional approaches.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1994년도 추계학술대회 논문집
• /
• pp.205-209
• /
• 1994

The main purpose of this study is constructing new velocity fields on the base of shape function used in finite element method and showing the possibility of application it to metal forming processes. Utilizing the 8-node quadratic rectangular element, we expressed the velocity within the deformation region by interpolating the velocity of each nodal points. And the upper-bound formulation from this velocity fields was derived. In order to confirm the validity of this method we applied it to axisymmetic combined extrusion problem. the results of load show that this method is on better agreement with experiment than the conventional UBET, and also the flow pattern and profile of extruded part are reasonable.

• 농업과학연구
• /
• 제40권2호
• /
• pp.169-176
• /
• 2013

The price and revenue of rice are expected to decrease due to increasing rice imports, decreasing consumption and the discontinuance of the government's rice procurement. This degenerating profitability is leading to a rise in the cultivation of upland-crops such as beans, fodder crops and fruits in paddy fields. However, there is a lack of research on the selection of rice substitute crops which are adaptable to the relevant region through profitability analysis. This research, therefore, analyzed investment profitability of rice substitute crops for Cheorwon-gun area in Kangwon province. The study applied net present value (NPV) and internal rate of return (IRR), which fit for mutually exclusive investments that make one selection to the exclusion of other crops. Target crops are green house plants in Cheorwon-gun area. Financial analysis showed paprika and cucumber have investment feasibility for automated vinyl greenhouses and conventional plastic greenhouses respectively.

• 한국정보처리학회논문지
• /
• 제6권5호
• /
• pp.1385-1392
• /
• 1999

본 논문에서는 웨이퍼를 세정하는데 사용되는 Wet Station 장비의 상태를 감지하고, 이를 간접적으로 제어하기 위한 모니터링 시스템의 설계에 관하여 기술한다. 대부분의 기존 모니터링 시스템은 하드웨어 및 소프트웨어에 의존되어 있다. 제시한 모니터링 시스템의 기본 설계 목표는 사용자의 편의성과 시스템의 이식성을 제공하는 것이다. 이러한 요구조건을 만족시키기 위하여, 본 논문에서는 IBM PC 호환 기종의 윈도우즈 NT 환경에서 GUL 기능과, TCP/IP 통신 프로토콜을 기본으로 한 EtherNet 보드를 이용하여 시스템을 설계하였다.

• 한국멀티미디어학회논문지
• /
• 제24권10호
• /
• pp.1319-1325
• /
• 2021

It is important to remove fog for accurate object recognition and detection during preprocessing because images taken in foggy adverse weather suffer from poor quality of images due to scattering and absorption of light, resulting in poor performance of various vision-based applications. This paper proposes an end-to-end deep learning-based single image de-fogging method using U-Net architecture. The loss function used in the algorithm is a loss function based on Mahalanobis distance with fog features, which solves the problem of domain shifts, and demonstrates superior performance by comparing qualitative and quantitative numerical evaluations with conventional methods. We also design it to generate fog through the VGG19 loss function and use it as the next training dataset.

• ETRI Journal
• /
• 제46권2호
• /
• pp.307-322
• /
• 2024

Most object detection methods use a horizontal bounding box that causes problems between adjacent objects with arbitrary directions, resulting in misaligned detection. Hence, the horizontal anchor should be replaced by a rotating anchor to determine oriented bounding boxes. A two-stage process of delineating a horizontal bounding box and then converting it into an oriented bounding box is inefficient. To improve detection, a box-boundary-aware vector can be estimated based on a convolutional neural network. Specifically, we propose a ResNeXt101 encoder to overcome the weaknesses of the conventional ResNet, which is less effective as the network depth and complexity increase. Owing to the cardinality of using a homogeneous design and multi-branch architecture with few hyperparameters, ResNeXt captures better information than ResNet. Experimental results demonstrate more accurate and faster oriented object detection of our proposal compared with a baseline, achieving a mean average precision of 89.41% and inference rate of 23.67 fps.

• 대한원격탐사학회지
• /
• 제36권6_2호
• /
• pp.1579-1590
• /
• 2020

본 논문에서는 고해상도의 위성영상을 활용하여 도시의 변화 양상을 분석하기 위하여 SPADE기반의 U-Net과 객체 영역기반 변화탐지 방법을 제안한다. 제안하는 네트워크는 기존의 U-Net에서 공간 정보를 잃지 않기 위해 SPADE를 사용했다. 고해상도 위성영상을 활용한 변화탐지 방법은 계획, 예측 등 다양한 도시 문제를 해결하기 위해 활용할 수 있다. IR-MAD 등 전통적인 방법인 화소 기반의 변화탐지를 수행할 경우, 다중 시기 영상 간의 기후, 계절 변화 등에 의해 화소의 변화가 민감하기 때문에 미변화 지역들이 변화 지역으로 오탐지될 가능성이 매우 크다. 이에 본 논문에서는 시계열 위성영상에서 도시를 구성하는 객체에 대한 변위를 정확하게 탐지하기 위해 도시를 구성하는 주요 공간 객체를 정의하고, 딥러닝 기반 영상 분할을 통해 추출한 후 영역 간의 변위 오차를 분석하여 변화탐지를 수행한다. 변화 양상을 분석하기 위한 공간 객체로 건축물, 도로, 농경지, 비닐하우스, 산림 영역, 수변 영역의 6개로 정의하였다. KOMPSAT-3A 위성영상으로 학습한 각 네트워크 모델을 시계열 KOMPSAT-3 위성영상에 대한 변화탐지를 수행한다. 객관적인 성능 평가를 위한 변화탐지 지표는 F1-score, Kappa를 사용한다. 제안하는 변화탐지 기법은 U-Net, UNet++ 대비 뛰어난 결과를 보이며, 평균 F1 score는 0.77, kappa는 77.29의 성능을 확인할 수 있다.

• Nuclear Engineering and Technology
• /
• 제46권1호
• /
• pp.27-38
• /
• 2014

The Fukushima Dai-Ichi accident in 2011 has affected various aspects of the nuclear society worldwide. The accident revealed some problems in the conventional approaches used to ensure the safety of nuclear installations. To prevent such disastrous accidents in the future, we have to learn from them and improve the conventional approaches in a more systematic manner. In this paper, we will cover three issues. The first is to identify the key issues that affected the progress of the Fukushima Dai-Ichi accident greatly. We examine the accident from a defense-in-depth point of view to identify such issues. The second is to develop a more systematic approach to enhance the safety of nuclear installations. We reexamine nuclear safety from a risk point of view. We use the concepts of residual and unknown risks in classifying the risk space. All possible accident scenarios types are reviewed to clarify the characteristics of the identified issues. An approach is proposed to improve our conventional approaches used to ensure nuclear safety including the design of safety features and the safety assessments from a risk point of view. Finally, we address some issues to be improved in the conventional risk assessment and management framework and/or practices to enhance nuclear safety.

• 한국유기농업학회지
• /
• 제26권4호
• /
• pp.629-647
• /
• 2018

1. 유기재배구의 토양조건은 배수가 약간 불량한 미사질 양토인 전북통 보통답으로 유효규산이 높은 것을 제외하고는 우리나라 논토양의 화학적 성분과 평균 함량이 유사하였다. 2. 관행재배구와 유기재배구 간에 생육은 초장, 분얼수, 이삭수, 이삭길이 모두 관행재배구에서 높은 것을 확인하였으며 이삭당 입수와 등숙율은 유기재배구에서 다소 높게 나타났다. 3. 병해충 발생정도는 물바구미, 잎집무늬마름병, 혹명나방, 도열병이 유기재배구에서 많이 발생하였고, 이삭누룩병은 관행재배구에서 조금 더 많이 발생하였다. 4. 주요 관리대상 병해충인 잎집무늬마름병(y=-0.2848x+199.57, $R^2=0.704^*$), 혹명나방 발생(y=-0.1361x+92.535, $R^2=0.8049^*$)과 쌀 수량과의 관계에서 유의성이 높은 부의 상관관계가 나타났다. 5. 백미 품질은 유기재배구에서 파쇄미의 증가로 완전립 비율이 낮았지만, 단백질함량이 관행재배구에 비해 낮아 전반적인 품질 및 윤기치가 높아졌다. 6. 쌀 재배 순수익은 관행재배구에서 36만원/10a, 유기재배구에서 61만원/10a로 유기재배구의 순수익이 68% 높게 나타났다.