• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2004.11a
• /
• pp.167-171
• /
• 2004

최근 들어 위성영상은 자료 처리 방식에 따라 지구표면이나 또는 지하면에 대한 다양한 정보(물리적인 정보, 화학적인 정보)를 얻을 수 있고 실제 지구를 가상으로 구현하는 데 활용될 수 있기 때문에 여러 산업에서 활용하고 있다. 또한 분류는 영상에 포함된 여러 가지 대상물을 구별하기 위해서 화소와 비교적 성질이 같은 화소 그룹별 특징에 대응되는 레벨을 지정하는 기술이 요구되며, 최소거리 분류법, 평행사변형법, 마하나로비스거리법(Mahanalobis Distance Method), 최대우도법(Maximum Likelihood Method)등 비교하여 분류를 수행

• Digital Contents
• /
• no.3 s.10
• /
• pp.84-94
• /
• 1994

센터내 DB표준화분과위원회 DB표준분류 실무작업반은 지난 93년도 하반기에 데이터베이스 표준분류를 위한 연구를 수행했다. 그간 실무작업반에서는 데이터베이스 분류에 관한 자료의 수집 및 비교분석, 데이터베이스의 제작기관, 주제분야별, 가공형태별, 표현형태별, 언어별, 가공완성도 및 갱신주기별, 검색방식별, 제공매체별, 용도별체계화 등 데이터베이스 표준분류안을 마련했는데 본지에서는 연구결과를 중심으로 그 내용을 정리, 요약한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.6
• /
• pp.799-803
• /
• 2007

In this paper, we propose a new fuzzy membership function for FSVM(Fuzzy Support Vector Machines). We apply a fuzzy membership to each input point of SVM and reformulate SVM into fuzzy SVM (FSVM) such that different input points can make different contributions to the learning of decision surface. The proposed method enhances the SVM in reducing the effect of outliers and noises in data points. This paper compares classification and estimated performance of SVM, FSVM(1), and FSVM(2) model that are getting into the spotlight in time series prediction.

• Proceedings of the KAIS Fall Conference
• /
• 2010.11b
• /
• pp.759-762
• /
• 2010

본 연구는 당뇨인지환자들의 당뇨 조절에 관계되는 요인들을 포괄적으로 반영하는 집단으로 분류한 후 이를 기반으로 보다 효율적인 당뇨관리사업을 할 수 있는 기초자료를 제공하기 위해 수행되었다. 연구를 위해 2007년, 2008년도 국민건강영양조사를 통해 검진에 참여한 당뇨인지환자 666명의 자료를 수집하여 분석하였다. 당뇨인지환자의 관리행태에 대한 군집분류는 K-means 기법을 이용하였다. 당뇨인지환자의 군집은 건강행태사업 대상군, 중점관리사업 대상군, 합병증검사사업 대상군으로 분류되었다. 당뇨 조절율을 높이기 위해서는 각 군집의 특성에 따라 보다 특화된 당뇨관리 프로그램이 적용되어야 할 것이다.

• Korean Journal of Environment and Ecology
• /
• v.13 no.2
• /
• pp.129-142
• /
• 1999

생태도시계획의 핵심토대로서 UBM(Urban Biotope Mapping)과 BIS(Biotope Information System)는 총 5단계로 구성된다. 본 연구는 2단계의 연구로서 독일 도르트문트시를 사례지로 한 도시비오톱 유형분류방법을 규명해 보고 우리 나라 적용가능성을 타진해 보는 데 목적이 있었다. 유형분류결과 biotope group complex는 총 12개, biotope group은 총 67개, detail biotope은 총 1,120개로 나타났다. 면적점유율 및 분포현황의 분석결과 습지초원 및 반건초지 초원비오톱(전체 면적 0.8%)이 희귀비오톱으로 분류되었으며, 분구원지역비오톱(전체 면적의 2.8%)의 다양도가 가장 높게 나타났다. 또한 적용가능성이란 측면에서 볼 때 서로 다른 도시발전형태에 따른 토지이용패턴의 상이성보완이 필요하고, 더불어 기타 도시생태관련자료에서 지형(적)도(1 : 5,000) 및 적외선 칼라항공사진(1 : 5,000)은 최소한의 자료로 준비되어야 할 것으로 사료되었다.

• Korean Journal of Remote Sensing
• /
• v.28 no.1
• /
• pp.21-28
• /
• 2012

In this study, detection of icebergs that have various scattering characteristics around Wilkinson glacier in West Antarctica is investigated using C-band fully-polarimetric RADARSAT-2 SAR data. Various polarimetric analyses including Freeman-Durden decomposition, H/A/$\bar{\alpha}$ decomposition, entropy (H) and anisotropy (A) method, and Wishart unsupervised classification, were applied for the RADARSAT-2 data used in this study. The polarimetric decomposition methods were successfully classified most of the iceberg, yet some iceberg with similar intensity of volume and surface scattering as sea ice were indistinguishable. Unsupervised classification with a combination of the polarimetric parameter, [1-H][1-A], gave a possibility to distinguish those unclassified iceberg.

• Journal of Environmental Impact Assessment
• /
• v.7 no.2
• /
• pp.103-112
• /
• 1998

Nature-friendly river project has became common practice in Japan. In order to make it available for the conservation and rehabilitation of desirable water environment, water criteria for water environmental assessment must be established. Especially, the criteria estimating the effects on ecosystem in and around river should be constructed. In this paper, classification method for water quality has been developed using biological indices and applied to observed data in Honmyo River, Nagasaki, Japan. Modified PI method (BI') has been suggested and those of three most abundant species resulted effective estimate for an overall water quality with comparatively simple procedure. Extensive database management code was prepared for the comprehensive ecological monitoring of river basin, which includes various biota. That system enables easy access of all the ecological data for a dissemination of a sound and sustainable water environment. The result of this study could improve knowledge base, serve making consensus for citizens, and help river management plans. In Japan, citizen's realization and action are the most critical factor for nature-friendly river restoration project.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.4
• /
• pp.471-476
• /
• 2010

This study aimed at monitoring, mapping, and assessing the land degradation in the desert area. In this research, the Landsat TM and ETM+ imageries to assess the extent of land degradation for study area during the period from 1991 to 2007. Were used to study supervized, unsupervized classfication and NDVI land cover changes in the desert area in Mongolia. The classified map consists of five classes of water, vegetation, slight desertification, middle desertification and sever desertification. It shows that for determination classfication methods and NDVI, desertification map of the study area are prepared. The result showed accounting for a clear deterioration in vegetative cover, an increase of sever desertification and a decrease in middle desertification and slight desertification respectively of the total study area.

• Korean Journal of Remote Sensing
• /
• v.33 no.5_2
• /
• pp.677-687
• /
• 2017

In this paper, we propose the use of hierarchical classification for winter crop mapping based on satellite imagery. A hierarchical classification is a classifier that maps input data into defined subsumptive output categories. This classification method can reduce mixed pixel effects and improve classification performance. The methodology are illustrated focus on winter cropsin Gimje city, Jeonbuk with Landsat-8 imagery. First, agriculture fields were extracted from Landsat-8 imagery using Smart Farm Map. And then winter crop fields were extracted from agriculture fields using temporal Normalized Difference Vegetation Index (NDVI). Finally, winter crop fields were then classified into wheat, barley, IRG, whole crop barley and mixed crop fields using signature from Unmanned Aerial Vehicle (UAV). The results indicate that hierarchical classifier could effectively identify winter crop fields with an overall classification accuracy of 98.99%. Thus, it is expected that the proposed classification method would be effectively used for crop mapping.

• The Korean Journal of Applied Statistics
• /
• v.28 no.3
• /
• pp.495-509
• /
• 2015

We study a classification problem of significant differences in the proportion of two groups known as the unbalanced classification problem. It is usually more difficult to classify classes accurately in unbalanced data than balanced data. Most observations are likely to be classified to the bigger group if we apply classification methods to the unbalanced data because it can minimize the misclassification loss. However, this smaller group is misclassified as the larger group problem that can cause a bigger loss in most real applications. We compare several classification methods for the unbalanced data using sampling techniques (up and down sampling). We also check the total loss of different classification methods when the asymmetric loss is applied to simulated and real data. We use the misclassification rate, G-mean, ROC and AUC (area under the curve) for the performance comparison.