• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.30-30
• /
• 2015

유역에 내린 강우의 총량은 홍수나 갈수 측면에서 매우 중요하다. 점 강우량에 의해 측정된 강우량을 이용하여 유역 총강우량으로 환산하는 과정에 많은 오차가 포함되어 있다. 선행연구에 따르면, 우량계를 통한 강우관측에서 언더캐치(undercatch)에 의한 계통오차는 일반적으로 5~16%, 우연오차는 약 5%가 발생된다고 보고하였으며, 점 우량계 자료를 내삽하여 공간자료로 변환할 경우 0.1km 규모에서 표준오차가 4~14%, 1km 규모에서는 33~45%, 10km 규모에서는 약 65% 정도 발생된다고 한다. 이러한 우량계 관측오차 및 강우자료 처리과정에서 발생되는 오차는 유역의 유출량 계산에 영향을 주어 홍수예보 정확도를 크게 떨어뜨릴 수 있다. 우리나라에서는 지금까지 유역 총강우량 산정 측면에서 지점강우량의 불확실성에 대한 연구가 많이 이루어지지 못하였다. 본 연구에서는 우리나라에서 주로 사용되고 있는 전도형 우량계를 이용하여 소규모 구역에서 관측되는 강우관측의 불확실성을 분석하고자 하였다. 연구에 사용된 우량계는 0.5mm 급 표준 전도형 우량계로 정밀도는 시간당 1~100mm 기준으로 ${\pm}1%$를 기록하여 기상검정규격인 ${\pm}3%$를 만족하고 있다. 이 우량계는 한국건설기술연구원 안동하천실험센터 내에 장애물이 없는 평지에 60m 간격으로 총 6대($2{\times}3$)를 설치하여 2014년 7월 11일부터 9월 2일까지 54일간 관측을 수행하였다. 관측기간 동안 2대의 우량계가 수일동안 강우가 기록되지 않아서 분석에서 제외하였다. 우량계 상호 간의 누적강우량(54일간)을 비교한 결과 2.5~25.5mm의 차이를 나타냈다. 강우강도별 강우량 합계를 비교한 결과 시간당 1mm 이상에서는 약 1%의 차이가 났으며, 시간당 15mm 이상에서는 7.4%의 차이를 나타내어 강도가 큰 강우사상에서 우량계 간의 관측오차가 더 크게 나타났다. 또한 우량계 상호 간의 상관계수를 분석한 결과, 우량계 간의 거리가 가까울수록 그리고 누적시간이 길수록 상관계수는 커지는 것을 확인할 수 있었다. 도출된 결과를 토대로 하면 앞서 언급한 바와 같이 점 우량계 자료를 내삽하거나 유역 또는 계산격자의 대푯값으로 사용하여 1시간 이하 단위로 유출모의를 할 경우 심각한 오차를 발생시킬 수 있음을 시사한다. 보다 신뢰성 있는 홍수예보와 효율적인 유역관리를 위해서는 점 중심의 강우 관측이 아닌 면적 우량에 대한 관측이 이루어져야 하며 이를 위한 기술의 개발이 필요하다.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.9
• /
• pp.67-76
• /
• 2010

This paper proposes a image-based water level measurement method, which adapt to the ruler's surface condition. When the surface of a ruler is deteriorated by mud, drifts, or strong light reflection, the proposed method judges the pollution of ruler by comparing distance between two levels: the first one is the end position of horizontal edge region which keeps the pattern of ruler's marking, and the second one is the position where the sharpest drop occurs in the histogram which is construct using image density based on the axis of image height. If the ruler is polluted, the water level is a position of local valley of the section having a maximum difference between the local peak and valley around the second level. If the ruler is not polluted, the water level is detected as the position having horizontal edges more than 30% of histogram's maximum value around the first level. The detected water level is converted to the actual water level by using the mapping table which is construct based on the making of ruler in the image. The proposed method is compared to the ultrasonic based method to evaluate its accuracy and efficiency on the real situation.

• The Korean Journal of Archival Studies
• /
• no.25
• /
• pp.91-129
• /
• 2010

As it grows more heavily reliant on information system, researchers seek for various ways to manage and utilize of dataset records which is accumulated in public information system. It might be needed to adjust date and enhance the quality of public administrative dataset records during transferring to archive system or sharing server. The purpose of this paper is presenting data adjustment and quality enhancement methods for public administrative dataset records, and it refers to ETT procedure and method of construction of datawarehouses. It suggests seven typical examples and processing method of data adjustment and quality enhancement, which are (1) verification of quantity and data domain (2) code conversion for a consistent code value (3) making component with combinded information (4) making a decision of precision of date data (5) standardization of data (6) comment information about code value (7) capturing of metadata. It should be reviewed during dataset record transfer. This paper made Data adjustment and quality enhancement requirements for dataset record transfer, and it could be used as data quality requirement of administrative information system which produces dataset.

• Journal of the Korean Society of Radiology
• /
• v.12 no.6
• /
• pp.793-799
• /
• 2018

As interest in artificial intelligence has increased, artificial intelligence has been actively studied in the medical field. In Korea, artificial intelligence has been applied to medical imaging devices such as X-ray imaging, Computer Tomography and Magnetic Resonance Imaging and artificial intelligence capable of acquiring radiation images of patients without radiologists in the future Medical devices are expected to be invented. This study was an initial study on the automation of patient positioning in X - ray imaging. We used x-ray equipment and human phantoms to evaluate the positioning. The program used Visual Studio 2010 MFC and the image was in the size $1450{\times}1814$. The pixel values were converted to contrasts with values of 0 to 255 that can be visually recognized and output to the monitor. We developed a procedure algorithm program that predicts the angle of the output image through three pixel coordinate values and induces the patient to perform correct positioning according to the voice guidance according to the angle. In the next study, we will study the artificial intelligence to grasp the structure itself and calculate the angle, rather than conveying the reference of coordinates to artificial intelligence. In the future, it is expected that it will be helpful in the study of artificial intelligence from shooting to positioning through the automation of positioning.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_3
• /
• pp.1383-1398
• /
• 2018

Sea surface wind is one of the most fundamental variables for understanding diverse marine phenomena. Although scatterometers have produced global wind field data since the early 1990's, the data has been used limitedly in oceanic applications due to it slow spatial resolution, especially at coastal regions. Synthetic Aperture Radar (SAR) is capable to produce high resolution wind field data. KOMPSAT-5 is the first Korean satellite equipped with X-band SAR instrument and is able to retrieve the sea surface wind. This study presents the validation results of sea surface wind derived from the KOMPSAT-5 backscattering coefficient data for the first time. We collected 18 KOMPSAT-5 ES mode data to produce a matchup database collocated with buoy stations. In order to calculate the accurate wind speed, we preprocessed the SAR data, including land masking, speckle noise reduction, and ship detection, and converted the in-situ wind to 10-m neutral wind as reference wind data using Liu-Katsaros-Businger (LKB) model. The sea surface winds based on XMOD2 show root-mean-square errors of about $2.41-2.74m\;s^{-1}$ depending on backscattering coefficient conversion equations. In-depth analyses on the wind speed errors derived from KOMPSAT-5 backscattering coefficient data reveal the existence of diverse potential error factors such as image quality related to range ambiguity, discrete and discontinuous distribution of incidence angle, change in marine atmospheric environment, impacts on atmospheric gravity waves, ocean wave spectrum, and internal wave.

• Journal of the Korean Society of Radiology
• /
• v.13 no.3
• /
• pp.481-487
• /
• 2019

Based on the scan conditions and algorithms that are generally applied during examinations during head CT examinations, the results of dose reduction through the application of algorithm changes were investigated through experiments. As a result, the dose reduction effect was more meaningful for the change of perfusion than for the tube voltage, and the quality evaluation using the brain phantom was relatively less reduced when the dose was reduced after the application of the Bone algorithm, especially for the application of the Bone algorithm, and the deviation of the mean CT number or Pixel value was measured relatively significantly. In other words, the conditions under which dose was reduced and quality was maintained to reduce the patient's exposure dose and obtain images of the same quality were obtained with the application of the Smooth algorithm and the resulting values of 120 kVp, 160 mA. At this point, doses were reduced by about 28%, and the mean CT number or Pixel value was also measured with relatively little error. If the results are applied to patients who visit the hospital for examination or follow-up after applying various algorithms and follow up scan conditions, the results are considered to be very useful in reducing patient exposure dose.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.4
• /
• pp.359-365
• /
• 2019

This paper is a basic study for the performance evaluation, durability improvement and optimal design of tractor transmission. The engine torque of the 78 kW agricultural tractor during rotary tillage was measured using CAN communication. It was calculated with equivalent torque and then analyzed. In order to develop a reliable tractor, it is important to convert measured torque in various agricultural operations into equivalent torque and analyze it. The equivalent torque was calculated using Palmgren-Miner's rule, which is a representative cumulative damage law. The equivalent torque of rotary tillage period and steering period are 229.2 and 136.7 Nm, respectively. The maximum and average torque during rotary tillage period are 336.0 and 234.4 Nm, respectively. The maximum and average torque of the steering period are 288.0 and 134.6 Nm, respectively. The engine torque in rotary tillage period is higher than in the steering period because of cultivation of soil through PTO. The maximum and rated torque of engine are 387.0 and 323.0 Nm, respectively, which are 183% and 136% higher than the equivalent torque during rotary tillage and of steering section. Because transmission of agricultural tractor in Korea companies is generally designed by the rated torque of engine, there is a difference from measured torque during agricultural operations. Therefore, it is necessary to consider it for optimal design.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.472-484
• /
• 2019

This paper proposes a time-domain sound event detection algorithm using DNN (Deep Neural Network). In this system, time domain sound waveform data which is not converted into the frequency domain is used as input to the DNN. The overall structure uses CRNN structure, and GLU, ResNet, and Squeeze-and-excitation blocks are applied. And proposed structure uses structure that considers features extracted from several layers together. In addition, under the assumption that it is practically difficult to obtain training data with strong labels, this study conducted training using a small number of weakly labeled training data and a large number of unlabeled training data. To efficiently use a small number of training data, the training data applied data augmentation methods such as time stretching, pitch change, DRC (dynamic range compression), and block mixing. Unlabeled data was supplemented with insufficient training data by attaching a pseudo-label. In the case of using the neural network and the data augmentation method proposed in this paper, the sound event detection performance is improved by about 6 %(based on the f-score), compared with the case where the neural network of the CRNN structure is used by training in the conventional method.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.51-56
• /
• 2019

This paper related with the performance comparison of CCA and RMMA blind adaptive equalization in order to reduce the intersymbol interference which is occurred in channel when transmitting the 16-QAM signal, high spectrum efficiencies of nonconstant modulus characteristic. The CCA possible to improve the misadustment and initial convergence by compacting the every signal constellation of 16 by using the sliced symbol of the decision device output, namely statistical symbol, but incresing the computational cost. The RMMA possible to minimize the fast convergence speed and misadjustment and channel tracking capability without increasing the computational cost by obtain the error signal after transform to 4 constant modulus signal based on the region of signal constellation located. In this paper, these algorithm were implemented in the same channel, and the blind adaptive equalization performance were compared using the equalizer output signal constellation, residual isi, MSE, SER. As a result of simulation, the RMMA has better performance in output signal constellation, residual isi and MSE compared to the CCA, but has slow convergence speed about 1.3 times. And the SER performance presenting the robustness to the noise signal, the CCA has more beeter in less SNR, but the RMMA has better in greater than 6dB in SNR.

• Journal of Internet Computing and Services
• /
• v.20 no.5
• /
• pp.87-94
• /
• 2019

In this paper, the method of analyzing the unformatted data of consumers accumulated on social networks in the era of the Fourth Industrial Revolution by utilizing data from the service design and social psychology aspects was proposed. First, the fandom phenomenon, which shows subjective and collective behavior in a space on a social network rather than physical space, was defined from a data service perspective. The fandom model has been transformed into a collective level of customer Persona that has been analyzed at a personal level in traditional service design, and social network analysis that analyzes consumers' big data has been presented as an efficient way to pattern and visually analyze it. Consumer data collected through social leasing were pre-processed by column based on correlation, stability, missing, and ID-ness. Based on the above data, the company's brand strategy was divided into active and passive interventions and the effect of this strategic attitude on the growth direction of the consumer's fandom community was analyzed. To this end, the fandom model of consumers was proposed by dividing it into four strategies that the brand strategy had: stand-alone, decentralized, integrated and centralized, and the fandom shape of consumers was proposed as a growth model analysis technique that analyzes changes over time.