• 한국지하수토양환경학회:학술대회논문집
• /
• 한국지하수토양환경학회 2003년도 총회 및 춘계학술발표회
• /
• pp.287-291
• /
• 2003

To delineate the internal structures of alluvial aquifer, high resolution seismic and GPR methods were adopted in Buyeo Gunsu-Ri area. The result of seismic refraction survey shows the water table of the aquifer and the result of seismic reflection reveals the basement and somewhat dominant internal structures of alluvial aquifer. The internal heterogeneity due to variations in channel behavior can be delineated using GPR survey. GPR profiles for the point bar deposits near Buyeo county reveals two different stratigraphic units the lower inclined heterogeneous strata and the upper horizontally stratified strata. According to the increase of demand for water resource using artificial recharge in alluvium, it is believed that the information acquired by high resolution geophysical methods will have an important roles for the effective and sustainable development and usage of groundwater in alluvial aquifer.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2008년도 추계 학술발표회
• /
• pp.660-671
• /
• 2008

The suspension bridge between Myodo and Gwangyang is located in the main navigation channel to Gwangyang Harbor. So, there is need for the collision protection against large vessels. As ship collision protection, artificial island with concrete block quay wall is planned. The risk analysis and non-linear numerical analysis are introduced to consider the ship collision effects. In the Gwangyang bay area, there are some different sedimental conditions in clayey stratums. For a desirable design, we classify into four zones and 2 layers in each zone, and then determine suitable soil properties considering these zones. As a ground improvement under artificial island, DCM and SCP methods are Planned.

• 한국멀티미디어학회논문지
• /
• 제25권1호
• /
• pp.52-60
• /
• 2022

Few-shot learning is a sub-area of machine learning problems, which aims to classify target images that only contain a few labeled samples for training. As a representative few-shot learning method, the Prototypical network has been received much attention due to its simplicity and promising results. However, the Prototypical network uses the sample mean of samples from the same class as the prototypes of that class, which easily results in learning uncharacteristic features in the low-data scenery. In this study, we propose to use local descriptors (i.e., patches along the channel within feature maps) from the same class to explicitly obtain more representative prototypes for Prototypical Network so that significant intra-class feature information can be maintained and thus improving the classification performance on few-shot learning tasks. Experimental results on various benchmark datasets including mini-ImageNet, CUB-200-2011, and tiered-ImageNet show that the proposed method can learn more discriminative intra-class features by the local descriptors and obtain more generic prototype representations under the few-shot setting.

• 대한토목학회논문집
• /
• 제38권2호
• /
• pp.227-235
• /
• 2018

본 연구에서는 차량용 레인센서를 이용하여 고해상도의 강우관측을 위한 관계식을 개발하였다. 차량용 레인센서는 8개의 채널로 이루어져 있으며, 각 채널은 강우발생시 차량의 전면유리창에 내리는 우적량을 감지하여 센서시그널을 생성하는데, 강우량이 높을수록 센서시그널은 낮은 값으로 형성된다. 레인센서에서 생성되는 센서시그널의 이러한 특징을 이용하여 관계식을 개발하였다. 특정강우를 발생시키기 위하여 인공강우 발생장치를 제작하였으며, 인공강우발생장치에서 분사되는 강우량의 변화에 따른 센서시그널의 변화 값을 분석하였다. 이 중 민감도 분석을 통해 다양한 강우량을 잘 반영하는 최적의 센서 채널을 선정하였다. 선정된 채널을 이용하여 5분 단위 센서시그널를 생성하였고 생성된 5분 단위 센서시그널의 대표 값을 평균, 25분위, 50분위, 75분위로 설정하여 관계식을 구축하였다. 구축된 관계식을 이용하여 실강우 데이터에 적용하여 강우량 값을 환산하였고, 환산된 강우량은 지상강우관측소에서 관측된 강우량 값과의 비교를 통해 관계식의 신뢰도를 검증하였다. 검증결과 데이터의 이상치가 발견되어 관계식의 신뢰도는 다소 떨어졌지만, 해당 잔차 범위의 실험 데이터가 부족한 것으로 판단되었다. 개발된 관계식을 실강우에 적용시켜 강우량 값을 환산한 하였고, 신뢰도 검증을 위해 동시간에 관측된 지상강우 관측 장비에서 생성된 강우량 값과 비교를 하였고 관측 결과 레인센서는 0.5mm 이하의 미세한 강우량까지 측정하였고 평균 관측 오차는 0.36mm로 나타났다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권2호
• /
• pp.631-653
• /
• 2020

Sign language is a natural, visually oriented and non-verbal communication channel between people that facilitates communication through facial/bodily expressions, postures and a set of gestures. It is basically used for communication with people who are deaf or hard of hearing. In order to understand such communication quickly and accurately, the design of a successful sign language translation system is considered in this paper. The proposed system includes object detection and classification stages. Firstly, Single Shot Multi Box Detection (SSD) architecture is utilized for hand detection, then a deep learning structure based on the Inception v3 plus Support Vector Machine (SVM) that combines feature extraction and classification stages is proposed to constructively translate the detected hand gestures. A sign language fingerspelling dataset is used for the design of the proposed model. The obtained results and comparative analysis demonstrate the efficiency of using the proposed hybrid structure in sign language translation.

• 수산해양기술연구
• /
• 제44권3호
• /
• pp.207-217
• /
• 2008

Costal regions in Korea often suffer severe damages due to wave-induced disasters, storm surge disasters and so on. therefore, many engineers and researchers have devoted their energy to prevent these costal disasters. The development of artificial reefs including sunken vessels is one of their remarkable achievements and various kind of these artificial upwelling structures have been designed and applied. However, the flow characteristics around a Tetrapod under the water has not been investigated experimentally. So in this article, in uniform flow of circulating water channel and some different velocities, PIV measurement has been conducted on the flow characteristics behind a Tetrapod. The results were analyzed on the flow characteristics of both cases of a Tetrapod. Therefore, it can be concluded that the both cases have its own distinctive flow characteristics behind the bluff body; Case A has an steep upstream flow pattern. On the contrary, Case B has an developed downstream flow pattern in the near wake of the Tetrapod. The velocity gradient at position x=150mm of Case-A appears gently up and down But, the velocity gradient at the same position of Case-B appears better highly up and down.

• Membrane and Water Treatment
• /
• 제13권5호
• /
• pp.219-225
• /
• 2022

5th Assessment Report of the Intergovernmental Panel on Climate Change Weather (AR5) predicts that recent severe hydrological events will affect the quality of water and increase water pollution. To analyze changes in water quality due to future climate change, input data (precipitation, average temperature, relative humidity, average wind speed, and solar radiation) were compiled into a representative concentration curve (RC), defined using 8.5. AR5 and future use are calculated based on land use. Semi-distributed emission model Calculate emissions for each target period. Meteorological factors affecting water quality (precipitation, temperature, and flow) were input into a multiple linear regression (MLR) model and an artificial neural network (ANN) to analyze the data. Extensive experimental studies of flow properties have been carried out. In addition, an Acoustic Doppler Velocity (ADV) device was used to monitor the flow of a large open channel connection in a wastewater treatment plant in Ho Chi Minh City. Observations were made along different streams at different locations and at different depths. Analysis of measurement data shows average speed profile, aspect ratio, vertical position Measure, and ratio the vertical to bottom distance for maximum speed and water depth. This result indicates that the transport effect of the compound was considered when preparing the hazard analysis.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2007년도 춘계학술대회 학술발표 논문집 제17권 제1호
• /
• pp.52-57
• /
• 2007

한정된 자원을 효율적으로 사용해야하는 이동통신망에서 멀티미디어 서비스 요구에 따른 무선 트래픽 채널을 할당하는 기법은 무선이라는 특수 환경으로 인해 제약을 받을 수밖에 없다. 이동망의 기지국의 경우 여러 무선 가입자 보드로부터 요구되는 서비스별 트래픽요구에 대한 채널 할당과 이에 대한 메인보드에서 처리해야 하는 작업 스케줄링은 무선과 CPU라는 서로 다른 환경을 잘 매핑하는 과제를 안고 있다. 본 논문에서는 음성과 데이터 호를 동시에 서비스하는 셀룰러 시스템에서 멀티미디어 서비스 트래픽 특성을 고려한 주파수할당과 작업 스케줄링이라는 두 가지 요소를 접목할 때 인공지능알고리즘인 유전자알고리즘을 이용하는 방법과 이에 적합한 작업 스케줄링 방식을 제안한다.

• Journal of information and communication convergence engineering
• /
• 제15권1호
• /
• pp.14-20
• /
• 2017

In this paper, we discuss physical layer security techniques in downlink networks, including eavesdroppers. The main objective of using physical layer security is delivering a perfectly secure message from a transmitter to an intended receiver in the presence of passive or active eavesdroppers who are trying to wiretap the information or disturb the network stability. In downlink networks, based on the random feature of channels to terminals, opportunistic user scheduling can be exploited as an additional tool for enhancing physical layer security. We introduce user scheduling strategies and discuss the corresponding performances according to different levels of channel state information (CSI) at the base station (BS). We show that the availability of CSI of eavesdroppers significantly affects not only the beamforming strategy but also the user scheduling. Eventually, we provide intuitive information on the effect of CSI on the secrecy performance by considering three scenarios: perfect, imperfect, and absence of eavesdropper's CSI at the BS.

• 유체기계공업학회:학술대회논문집
• /
• 유체기계공업학회 2001년도 유체기계 연구개발 발표회 논문집
• /
• pp.76-80
• /
• 2001

Unsteady compressible Euler equation is solved and the high-order, high-resolution numerical solver, physical boundary condition, adaptive nonlinear artificial dissipation model and conformal mapping are applied to computation of steady transonic flow and unsteady acoustics. The acoustic characteristics of axi-symmetric duct and two dimensional straight/S channel are studied and the computation results shows good agreements with linear analysis. In transonic case, local time stepping and canceling-the-residual techniques are used for convergence acceleration. The aspect of flow and acoustics in S-channel and the Pattern of noise radiation is changed by inflow Mach no. and static pressure at fan-face.