• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.404-404
• /
• 2011

최근 기상변동성 증가와 극치수문사상의 발생빈도 증가로 인한 기상재해가 빈번하게 일어나고 있다. 이러한 기상현상으로 인한 재해의 예방을 위해서 사전에 위험을 인지하고 그 규모를 예측할 수 있는 여러 기법들이 기상레이더 또는 수치예보자료 등을 이용하여 개발 및 적용되고 있다. 이 과정에서 해결해야 할 여러 문제점들이 있는데, 우선 수치예보자료 또는 기상레이더자료를 종관기상관측소 및 자동기상관측지점의 지상관측 강수량과 연계하여 평가하는 과정이 필요하고, 현재시점에 형성되어 있는 강우장의 공간 이동 예측 기법이 확보되어야 할 것이다. 전북지역은 게릴라성 집중호우가 빈번한 산악형 강수와 산지유역의 급한 하천경사가 맞물려 인명 및 재산피해가 매년 발생하고 있으며, 과거 돌발홍수가 발생한 사례가 있어 이상기후 및 기후변화로 인한 홍수 위험도가 커질 것으로 전망되고 있다. 본 연구는 전라북도의 기상재해 예측모형 개발을 위한 사전 분석과정으로 전라북도지역에서 관측된 기존의 대규모 강수사상을 이용한 강수사상의 특성 분류 및 관측소간 공간상관성을 분석하는데 목적을 두고 있다. 강수사상의 특성분류를 통해 강수 발생형태에 따른 기상학적 영향인자, 강수의 발생량 및 이동특성 예측의 정도를 향상시킬 수 있으며, 분류 기법으로 SVM(support vector machine)을 이용한 자동분류를 적용한다. 또한 관측소간 공간상관성 분석을 위하여 각 관측소 강수량간의 조건부 확률을 이용한다. 예로써 부안관측소에 강수가 발 생했을 때, 부안관측소의 강수량 조건에 의한 전주관측소 강수량 확률을 다음과 같이 구성할 수 있다. �揚滑斂�수량�咀刮활�수량��. 공간상관성 분석과정에서 관측소간 강수 이동시간에 따른 강수 발생 시간의 차이 또한 고려하며, 과거 기상관측 자료의 분석을 통해 전라북도지역의 관측소간 강수발생의 공간적 상관성을 규명하고, 단기예측 모델 개발을 위한 기초자료로 활용할 수 있을 것이다. 또한, 기후변화시나리오에 의한 미래 강수량의 지역적 상세화 과정에도 본 연구를 통한 결과를 이용할 수 있을 것이라 판단된다.

• Korean Journal of Remote Sensing
• /
• v.22 no.2
• /
• pp.101-110
• /
• 2006

The accuracy of rural land cover using airborne multispectral images and LEAR (Light Detection And Ranging) data was analyzed. Multispectral image consists of three bands in green, red and near infrared. Intensity image was derived from the first returns of LIDAR, and vegetation height image was calculated by difference between elevation of the first returns and DEM (Digital Elevation Model) derived from the last returns of LIDAR. Using maximum likelihood classification method, three bands of multispectral images, LIDAR vegetation height image, and intensity image were employed for land cover classification. Overall accuracy of classification using all the five images was improved to 85.6% about 10% higher than that using only the three bands of multispectral images. The classification accuracy of rural land cover map using multispectral images and LIDAR images, was improved with clear difference between heights of different crops and between heights of crop and tree by LIDAR data and use of LIDAR intensity for land cover classification.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.26 no.3
• /
• pp.238-247
• /
• 2021

Diatoms are a type of microalgae which inhabits a wide variety of environments as one of the most important primary producers of freshwater and marine ecosystems. They have been widely used as bioindicators which represent the environmental characteristics, thus proper quality control of diatom data is very important to ensure for the researches from many scientists from various different regions to have scientific unity and objectivity. At present, diatom data analysis is primarily based both on morphological features and DNA sequences of given species. Challenge for the morphology-based analysis of diatoms is the consistent species identification among different taxonomists who interpret diatom community, while challenge for the molecular analysis of diatoms to secure reliable reference data. In the present study, we have reviewed the current status of data quality control of diatom analysis in Korea as well as the world. Finally, suggestions for the better data quality control for Korean marine diatoms have been also made.

• The Korean Journal of Applied Statistics
• /
• v.32 no.3
• /
• pp.349-374
• /
• 2019

A class imbalance problem arises when one class outnumbers the other class by a large proportion in binary data. Studies such as transforming the learning data have been conducted to solve this imbalance problem. In this study, we compared resampling methods among methods to deal with an imbalance in the classification problem. We sought to find a way to more effectively detect the minority class in the data. Through simulation, a total of 20 methods of over-sampling, under-sampling, and combined method of over- and under-sampling were compared. The logistic regression, support vector machine, and random forest models, which are commonly used in classification problems, were used as classifiers. The simulation results showed that the random under sampling (RUS) method had the highest sensitivity with an accuracy over 0.5. The next most sensitive method was an over-sampling adaptive synthetic sampling approach. This revealed that the RUS method was suitable for finding minority class values. The results of applying to some real data sets were similar to those of the simulation.

• The KIPS Transactions:PartD
• /
• v.19D no.2
• /
• pp.133-138
• /
• 2012

Queries entered in a search box are the results of users' activities to actively seek information. Therefore, search logs are important data which represent users' information needs. The purpose of this study is to examine if there is a relationship between the results of queries automatically classified and the categories of documents accessed. Search sessions were identified in 2009 NDSL(National Discovery for Science Leaders) log dataset of KISTI (Korea Institute of Science and Technology Information). Queries and items used were extracted by session. The queries were processed using an automatic classifier. The identified queries were then compared with the subject categories of items used. As a result, it was found that the average similarity was 58.8% for the automatic classification of the top 100 queries. Interestingly, this result is a numerical value lower than 76.8%, the result of search evaluated by experts. The reason for this difference explains that the terms used as queries are newly emerging as those of concern in other fields of research.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.3 no.3
• /
• pp.163-170
• /
• 2001

The objective of this study was to evaluate land cover classification of PyeongTaeg area by Landsat Thematic Mapper Data June, 1997. This study was also to make more correct reference data using DGPS, aerophoto, and topographical chart etc.. The result of the area of paddy and upland were estimated 4,949 $\textrm{km}^2$ and 16,157 $\textrm{km}^2$, respectively. Correctness of estimation by using DGPS, aerophoto, topographical chart were shown over 90% correct in case of rice paddy field, water, and sea, while upland, vinyl house, forest, grassland, village were shown low correctness. Total average accuracy was shown to be 85.8%. Correctness of paddy field showed high value of 92%, showing that use of remote sensing data was proved to be effective methods to estimate spatial distribution and cultivation status of paddy field. Classification result of sea, water area, downtown had higher correctness, while upland, vinyl-house, grassland were proved to be relatively low correctness because of it's small area and mixed distribution.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.7 no.2
• /
• pp.85-92
• /
• 1989

The aim of this study is to find out the combination of effective transformed data, applying Remote Sensing techniques, as to the classification and particular objects by transforming the MSS data and TM data of the satellite LANDSAT into several linearly transformed data. Since one of the problems in the processing of the LANDSAT data is the vastness of the data, the Linear Transformation could be a method to perform analysis of those vast data, more efficiently and economically. This method is carried out as follows : (1) offering the simplicity over complex data, (2) selectional processing over redundant data and removing unnecessary data, (3) emphasizing on the object of the study ; by transforming multispectral data through linear calculation and statistical transformation. In this study, the analysis and transformation of the data have been performed by means of Band Ratioing and Principal Component Analysis. As the classificatory consequence, Infrared/RED Ratioing which expands the characterization of green vegetation, has been useful for a distinctive classification among other classes. For the Principal Component Analysis, band 1,2,7 are efficient in the classification of the green vegetation.

• 한국정보교육학회:학술대회논문집
• /
• 2004.08a
• /
• pp.577-586
• /
• 2004

인터넷을 통해 전달되는 교육자료의 양은 나날이 증가하고 있지만 정보 생산자들이 일정한 표준 없이 자의적인 기준에 의해 정보를 분류하여 구성하기 때문에 이용자가 필요한 정보를 정확하게 찾아내기란 매우 어려운 실정이다. 따라서 털 연구는 국내 주제별 검색엔진인 Yahoo Korea와 Naver, Hanmir, Empas의 초등학생용 검색엔진의 분류체계를 비교 분석하여 주제별 검색을 위한 효율적인 카테고리 분류 방법을 제시함으로써 정보접근에 익숙하지 않은 초등학생에게 쉽게, 빠르게, 정확하게 교육자료에 접근할 수 있는 분류체계를 제시하고자 한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.322-322
• /
• 2011

지표에서의 토양수분은 작은 구성비를 가짐에도 불구하고 여러 수문 현상을 연계하는 매우 중요한 인자로써 최근 연구가 활발하게 진행되고 있다. 토양수분은 침투나 침루를 통하여 강우와 지하수를 연결하는 기능을 함과 동시에 강우사상에 따른 유출특성에 직접적인 영향을 미치며 증발산을 통하여 에너지 순환을 연결하는 기능을 하는 인자로 기후변화와 인간의 활동에 의해 영향을 받는다. 지난 수십 년간 산림개간과 도시화는 토지이용의 변화를 초래하여 토지피복의 변화를 초래하였다. 도시화는 불투수층을 증가시켰고, 산림개간으로 산림이 농장으로 변하여 침투율을 감소시켜 유출률의 증가를 초래하였다. 이처럼 토지피복의 변화는 토양수분의 변화에 직접적인 영향을 미친다. 본 연구에서는 토지피복 분류를 위해 구름의 영향이 적은 Landsat TM 영상을 사용하여 청미천 유역의 토지피복을 분류하여 토지피복도를 작성하였다. 청미천 유역은 현재 국제수문관측사업(IHP)의 일환으로 체계적인 수문관측이 진행되고 있는 지점으로, 추후 인공위성 영상을 통해 산정한 토양수분 자료를 비교할 수 있는 유역이다. Landsat TM 영상은 2009년 5월 23일에 관측된 115-34(path row) 영상으로 구름이 거의 없는 날의 자료를 사용하였다. 다중 스펙트럴 위성영상인 Landsat TM 영상은 30m 공간해상도로써 토지피복분류와 식생 등의 정보를 추출하는데 적합한 것으로 알려져 있다. 청미천 유역의 위성영상에 대하여 영상의 전처리 과정을 거쳐 무감독분류와 감독분류기법을 적용하여 토지피복을 분류하였다. 분류한 토지피복도는 국토해양부에서 국가수자원관리 종합정보시스템(WAMIS) 을 통하여 제공되는 토지피복도와 비교하였다.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_2
• /
• pp.1239-1249
• /
• 2018

Optical remote sensing sensors provide visually more familiar images than radar images. However, it is difficult to discriminate sea ice types in optical images using spectral information based machine learning algorithms. This study addresses two topics. First, we propose a semantic segmentation which is a part of the state-of-the-art deep learning algorithms to identify ice types by learning hierarchical and spatial features of sea ice. Second, we propose a new approach by combining of semi-supervised and active learning to obtain accurate and meaningful labels from unlabeled or unseen images to improve the performance of supervised classification for multiple images. Therefore, we successfully added new labels from unlabeled data to automatically update the semantic segmentation model. This should be noted that an operational system to generate ice type products from optical remote sensing data may be possible in the near future.