• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.226-226
• /
• 2017

최근 국지성 집중호우 및 돌발홍수와 같은 급격한 기상변화로 인한 기상재해의 발생빈도가 증가함에 따라 고해상도의 기상레이더 강수자료를 사용한 수공학 분야의 연구가 활발하게 진행되고 있다. 레이더 강수자료를 수문분석에 활용하는 목적은 레이더 강수량이 제공하는 공간분포를 최대한 활용하는데 있다. 기상레이더는 광범위한 영역에 대하여 시공간적으로 연속적인 관측이 가능하므로 지상 강수자료에 비하여 고해상도의 강수자료를 확보하는데 이점이 있다. 본 연구에서는 고해상도의 레이더 강수자료의 공간분포 특성을 유지하면서 지상 강수자료의 양적특성을 유지할 수 있는 조건부 합성기법을 개발하였다. 레이더 강수자료와 지상 강수자료를 조건부 합성하기 위하여 널리 활용되고 있는 Kriging, 역거리 가중법 및 Spline 보간법을 적용하였다. 조건부 합성결과는 지상 강수패턴을 현실성 있게 재현하였다. 추가적으로 미계측 지점으로 간주하여 보간법에 적용되지 않은 강수자료와 조건부 합성기법 결과에 대하여 교차검증을 수행한 결과 조건부 합성기법을 통한 강수정보는 수문분석에 직접적으로 활용될 수 있는 가능성을 확인하였다. 본 연구결과를 향후 초단기 레이더 강수예측기법과 연계하여 수문모형의 입력 자료로 활용한다면 보다 진보된 수문해석이 가능할 것으로 판단된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.04a
• /
• pp.382-385
• /
• 2012

이 논문에서는 영상 검색(image retrieval) 및 영상 부류(image categorization)을 위하여 영상을 기술할 때 영상의 클레스(class)별로 서로 다른 주요 특징량(feature)에 가중치 를 주는 방법론을 제안한다. 기존에 연구되어온 영상의 특징량 벡터에 가중치를 주는 방식은 모든 영상 클레스에 대하여 동일하게 가중치를 적용하기 때문에 영상이 클레스별로 서로 다른 특징량이 중요하다는 성질을 이용할 수 없다. 영상이 클레 별로 서로 다른 특징량이 중요하다는 성질을 이용하기 위하여 영상의 클레스별로 특징량 벡터에 서로 다른 가중치 벡터(weight vector)를 학습하였다. 그 후 질의 영상(query image)이 입력되면, 기존의 영상 검색 프레임워크(framework)를 통해 데이터베이 스(database)로 부터 미리 정의된 서브 클레스(sub-class)의 수에 해당하는 영상부 집합(subset)을 만들었다. 그리고 영상부 집합의 특징량 벡터들에 클레스별로 각각 학습된 가중치 벡터를 적용하여 특징량 벡터들 간의 거리를 다시 계산하여 리랭킹(re-ranking)하였다. 이 방법론을 UKBench Dataset에 적용하여 실험을 해보았으며 가중치를 주기 전과 비교 하였을 때 더 높은 정확도를 보였다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.6
• /
• pp.1059-1068
• /
• 2021

In urban areas, GPS(Global Positioning System)/GNSS(Global Navigation Satellite System) signals are blocked or distorted by structures such as buildings, which limits positioning. To compensate for this problem, in this paper, fingerprinting-based positioning using RSRP(: Reference Signal Received Power) information of LTE signals is performed. The W-KNN(Weighted - K Nearest Neighbors) technique, which is widely used in the positioning step of fingerprinting, yields different positioning performance results depending on the similarity distance calculation method and weighting method used in correlation. In this paper, the performance of the fingerprinting positioning according to the techniques used in correlation is comparatively analyzed experimentally.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.446-446
• /
• 2022

본 연구에서는 MODIS(MODerate resolution Imaging Spectroradiometer) 다중 위성영상을 기반으로 전국 시공간 토양수분 및 토양수분 기반의 가뭄지수 SWDI(Soil Water Deficit Index)를 산정하였다. 시공간 토양수분의 산정을 위해 입력자료로 MODIS 위성의 지표면온도(Land Surface Temperature, LST), 증발산 및 식생(Enhanced Vegetation Index, EVI; Fraction of Photosynthetically Active Radiation, FPAR; Leaf Area Index, LAI; Normalized Difference Vegetation Index, NDVI) 관련 산출물 자료와 지상 관측자료인 일 단위 강수량 자료를 구축하였다. MODIS 위성영상은 산출물별로 제공되는 QC(Quality Control) 영상을 활용해 보정을 수행하였고, 공간 강수량 자료는 기상청에서 제공하는 전국 92개 지점의 종관기상관측자료를 구축하여 공간보간기법인 역거리가중법을 적용해 생성하였다. 실측 토양수분은 농촌진흥청에서 제공하는 76개 지점의 토양 깊이 10 cm에 설치된 TDR(Time Domain Reflectomerty) 센서에서 측정된 토양수분 자료를 활용하였으며, 토양수분 모의 시 토양 속성을 고려하기 위해 국립농업과학원에서 제공하는 토양도를 구축하여 활용하였다. 토양수분 산정 모형은 다중선형회귀모형(Multiple Linear Regression Model, MLRM)을 활용하였으며, 계절 및 토성에 따른 회귀식을 산정하였다. 회귀식 기반의 토양수분과 토성별 포장용수량 및 영구위조점 값을 이용하여 SWDI를 산정하고, 실제 가뭄 발생 시기 및 지역과의 비교하고자 한다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.6
• /
• pp.91-102
• /
• 2020

This study constructed production frontiers of inputs and outputs in a sequential manner, measured inefficiencies by applying a non-radial sequential weighted Russell directional distance function to these frontiers, and analyzed Luenberg productivity indices and the contribution of each of input and output factor based on these distances. The results are as follows. First, the productivity of banks increased due to technical changes after the global financial crisis. Second, productivity growth decreased between 2009 and 2014 due to technical changes after the recession, as previous studies have shown that technology progressed before the global financial crisis but then largely decreased or remained the same thereafter. After 2014, the productivity of banks improved. This result may be due to both technology improvement after 10 years of stagnation and reduction of inputs and non-performing loans. Third, the 3.6% annual of productivity growth for 10 years was comprised of 1.77% household loans, 0.67% corporate loans, 0.98% manpower, 1.18% non-performing loans, -0.5% total deposits, and -1.25% securities. Finally, this study has limitations since it could not control risks such as capital structure and interest volatility.

• Journal of Korea Water Resources Association
• /
• v.53 no.10
• /
• pp.903-915
• /
• 2020

Long-term high-resolution hydro-meteorological data has been recognized as an essential element in establishing the water resources plan. The increasing demand for spatial precipitation in various areas such as climate, hydrology, geography, ecology, and environment is apparent. However, potential limitations of the existing area-weighted and numerical interpolation methods for interpolating precipitation in high altitude areas remains less explored. The proposed PRISM (Precipitation-Elevation Regressions on Independent Slopes Model) model can produce gridded precipitation that can adequately consider topographic characteristics (e.g., slope and altitude), which are not substantially included in the existing interpolation techniques. In this study, the PRISM model was optimized with SCEM-UA (Shuffled Complex Evolution Metropolis-University of Arizona) to produce daily gridded precipitation. As a result, the minimum impact radius was calculated 9.10 km and the maximum 34.99 km. The altitude of coastal weighted was 681.03 m, the minimum and maximum distances from coastal were 9.85 km and 38.05 km. The distance weighting factor was calculated to be about 0.87, confirming that the PRISM result was very sensitive to distance. The results showed that the proposed PRISM model could reproduce the observed statistical properties reasonably well.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.9
• /
• pp.351-360
• /
• 2018

The LWR (Locally Weighted Regression) model, which is traditionally a lazy learning model, is designed to obtain the solution of the prediction according to the input variable, the query point, and it is a kind of the regression equation in the short interval obtained as a result of the learning that gives a higher weight value closer to the query point. We study on an incremental ensemble learning approach for LWR, a form of lazy learning and memory-based learning. The proposed incremental ensemble learning method of LWR is to sequentially generate and integrate LWR models over time using a genetic algorithm to obtain a solution of a specific query point. The weaknesses of existing LWR models are that multiple LWR models can be generated based on the indicator function and data sample selection, and the quality of the predictions can also vary depending on this model. However, no research has been conducted to solve the problem of selection or combination of multiple LWR models. In this study, after generating the initial LWR model according to the indicator function and the sample data set, we iterate evolution learning process to obtain the proper indicator function and assess the LWR models applied to the other sample data sets to overcome the data set bias. We adopt Eager learning method to generate and store LWR model gradually when data is generated for all sections. In order to obtain a prediction solution at a specific point in time, an LWR model is generated based on newly generated data within a predetermined interval and then combined with existing LWR models in a section using a genetic algorithm. The proposed method shows better results than the method of selecting multiple LWR models using the simple average method. The results of this study are compared with the predicted results using multiple regression analysis by applying the real data such as the amount of traffic per hour in a specific area and hourly sales of a resting place of the highway, etc.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.31 no.4
• /
• pp.283-293
• /
• 2015

Purpose: The aims of this study are to analyze and to visualize distribution of patients visiting at a dental college hospital, using geographic information system (GIS). The visualized data can be utilized in patient care and treatment planning, ultimately leading to the assessment of risk evaluation and prevention of dental diseases. Materials and Methods: Patient information data was obtained from Dankook University Dental Hospital including the unit number, gender, date of birth, and address from 2007 to 2014. Patient distribution was visualized using GIS. Statistical analyses were performed using SAS 9.3 and ArcGIS 10.1. Five factors including proximity, accessibility, age, gender, and socioeconomic status were investigated as the explanatory variables of the patient distribution. Results: The visualized patient data showed a nationwide scale of the patient distribution. There was a little difference in characteristics for each department. As closer at Dankook University Dental Hospital, visitors increased. And it strongly showed that elderly patients in rural areas tend to visit more. Conclusion: The distribution of patients has been shown to be significantly affected by the proximity, accessibility, age, gender and socioeconomic status. The underlying reason remains to be further studied.

• Journal of Korean Society for Geospatial Information Science
• /
• v.18 no.3
• /
• pp.37-48
• /
• 2010

The purpose of this paper is to evaluate how the rainfall field effect on a runoff simulation using grid radar rainfall data and ground gauge rainfall. The Gwangdeoksan radar and ground-gauge rainfall data were used to estimate a spatial rainfall field, and a hydrologic model was used to evaluate whether the rainfall fields created by each method reproduced a realistically valid spatial and temporal distribution. Pilot basin in this paper was the Naerin stream located in Inje-gun, Gangwondo, 250m grid scale digital elevation data, land cover maps, and soil maps were used to estimate geological parameters for the hydrologic model. For the rainfall input data, quantitative precipitation estimation(QPE), adjusted radar rainfall, and gauge rainfall was used, and then compared with the observed runoff by inputting it into a $Vflo^{TM}$ model. As a result of the simulation, the quantitative precipitation estimation and the ground rainfall were underestimated when compared to the observed runoff, while the adjusted radar rainfall showed a similar runoff simulation with the actual observed runoff. From these results, we suggested that when weather radars and ground rainfall data are combined, they have a greater hydrological usability as input data for a hydrological model than when just radar rainfall or ground rainfall is used separately.

• Journal of Cadastre & Land InformatiX
• /
• v.48 no.1
• /
• pp.169-186
• /
• 2018

The social influence of the elderly population will accelerate in a rapidly aging society. The purpose of this study is to establish a methodology for extracting vulnerable districts of the welfare of the aged through machine learning(ML), artificial neural network(ANN) and geospatial analysis. In order to establish the direction of analysis, this progressed after an interview with volunteers who over 65-year old people, public officer and the manager of the aged welfare facility. The indicators are the geographic distance capacity, elderly welfare enjoyment, officially assessed land price and mobile communication based on old people activities where 500 m vector areal unit within 15 minutes in Yongin-city, Gyeonggi-do. As a result, the prediction accuracy of 83.2% in the support vector machine(SVM) of ML using the RBF kernel algorithm was obtained in simulation. Furthermore, the correlation result(0.63) was derived from ANN using backpropagation algorithm. A geographically weighted regression(GWR) was also performed to analyze spatial autocorrelation within variables. As a result of this analysis, the coefficient of determination was 70.1%, which showed good explanatory power. Moran's I and Getis-Ord Gi coefficients are analyzed to investigate spatially outlier as well as distribution patterns. This study can be used to solve the welfare imbalance of the aged considering the local conditions of the government recently.