• Journal of the Korea Computer Graphics Society
• /
• v.15 no.3
• /
• pp.19-26
• /
• 2009

In this paper, a new adaptive vertex clustering using a KD-tree is presented for 3D mesh partitioning. A vertex clustering is used to divide a huge 3D mesh into several partitions for various mesh processing. An octree-based clustering and K-means clustering are currently leading techniques. However, the octree-based methods practice uniform space divisions and so each partitioned mesh has non-uniformly distributed number of vertices and the difference in its size. The K-means clustering produces uniformly partitioned meshes but takes much time due to many repetitions and optimizations. Therefore, we propose to use a KD-tree to efficiently partition meshes with uniform number of vertices. The bounding box region of the given mesh is adaptively subdivided according to the number of vertices included and dynamically determined axis. As a result, the partitioned meshes have a property of compactness with uniformly distributed vertices.

• Journal of The Korean Association For Science Education
• /
• v.42 no.6
• /
• pp.611-619
• /
• 2022

The purpose of this study is to explore the possibility of applying big data analysis to provide appropriate feedback to students using evaluation data in science education at a time when interest in educational data mining has recently increased in education. In this study, we use the evaluation data of 2,576 students who took 24 questions of the national assessment of educational achievement. And we use K-means cluster analysis as a method of unsupervised machine learning for clustering. As a result of clustering, students were divided into six clusters. The middle-ranking students are divided into various clusters when compared to upper or lower ranks. According to the results of the cluster analysis, the most important factor influencing clusterization is academic achievement, and each cluster shows different characteristics in terms of content domains, subject competencies, and affective characteristics. Learning motivation is important among the affective domains in the lower-ranking achievement cluster, and scientific inquiry and problem-solving competency, as well as scientific communication competency have a major influence in terms of subject competencies. In the content domain, achievement of motion and energy and matter are important factors to distinguish the characteristics of the cluster. As a result, we can provide students with customized feedback for learning based on the characteristics of each cluster. We discuss implications of these results for science education, such as the possibility of using this study results, balanced learning by content domains, enhancement of subject competency, and improvement of scientific attitude.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.26 no.6
• /
• pp.633-641
• /
• 2008

This study introduces the method of detecting and restoring occlusion areas by using epipolar algorithm and K-means classification algorithm for true ortho-photo generation. In the past, the techniques of detecting occlusion areas are using the reference images or information of buildings. But, in this study the occlusion areas can be automatically detected by using DTM data and exterior orientation parameters. The detected occlusion areas can be restored by using anther images or the computed values which are determined in K-means classification algorithm. In addition, this method takes advantages of applying epipolar algorithm in order to find same location in overlapping areas among images.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.481-483
• /
• 2018

AR은 VR과 달리 실세계 공간의 객체에 대한 서비스를 제공하므로 서비스 개발을 방해하는 많은 요인들이 발생한다. 이를 보완하기위해 비주얼 마커, SLAM, 객체인식 등 여러 AR 기술이 존재한다. 본 논문은 AR 기술 중에서 객체인식의 정확도 향상을 위해 지역가변 이진화(Local variable binarization)와 색상의 군집화를 사용해서 이미지에서 객체를 추출하는 방법을 제안한다. 지역 가변화는 픽셀을 순차적으로 읽어 들이면서 픽셀 주위의 값의 평균을 구하고, 이 값을 해당 픽셀의 임계 값으로 사용하는 알고리즘이다. 픽셀마다 주위 색상 값에 의해 임계 값이 변화되므로 윤곽선 표현이 기존의 이진화보다 뚜렷이 나타난다. 색상의 군집화는 객체의 중요색상과 배경의 중요색상을 중심으로 유사한 색상끼리 군집화 하는 것이다. 객체 내에서 가장 많이 나온 값과 객체 외에 가장 많이 나온 값을 각 각 기준으로 색조와 채도의 값을 Euclidean 거리를 사용해 객체의 색상과 배경 색상을 분리했다.

• International Journal of Highway Engineering
• /
• v.7 no.4 s.26
• /
• pp.211-219
• /
• 2005

A selection of the analysis section for traffic accidents is used to analyze definitely the cause of accidents sorting similar accidents by a group and to raise the effect of improvement projects deciding the priority of accidents. In the existing method, an uniformly dividing method based on road mileages has been used, which has no consideration for similarities among accidents. Consequently, in recent, a slider-length method considering accident types rather than road mileages is widely used. In this study, using K-mean method, a non-hierarchical grouping technique used in the Cluster Analysis ai a applicatory method for the slider length method, a method classifies accidents that occurred the most nearby mileages into one group is proposed. To verify the proposed method, a comparison between the f-mean method and the dividing method at regular intervals on the data of a total of 25.6km lengths along Kyung-bu freeway in Pusan direction was made so that the K-mean method was proved to an effective method considering the similarities and adjacencies of accidents.

• Journal of Korean Society of Forest Science
• /
• v.103 no.1
• /
• pp.51-58
• /
• 2014

Genetic diversity and genetic structures were estimated in seven natural populations of Phellodendron amurense Rupr in South Korea using ISSR markers. The average of polymorphic loci per primer and the proportion of polymorphic loci per population were 4.5 and 78.8% respectively with total 27 polymorphic loci from 6 ISSR primers. The Shannon's diversity index(I) was 0.421 and the expected heterozygosity($H_e$) was 0.285, which was similar to the heterozygosity (hs =0.287) inferred by Bayesian method. In AMOVA, 7.6% of total genetic variation in the populations was resulted from the genetic difference among populations and the other 92.4% was resulted from the difference among individuals within populations. Genetic differentiation(${\theta}^{II}$) and inbreeding coefficient(f) for total population were estimated to be 0.066 and 0.479 by Bayesian method respectively. In Bayesian clustering analysis, seven populations were assigned into three groups. This result was similar to the results of genetic relationships by UPGMA and PCA. The first group included Hwachoen, Gapyeong, Bongpyeong and Yongpyeong population, and the second included two populations in Sancheong region. Muju population was discretely assigned into the third group in spite of the geographically short distance from the Sancheong region. There was no significant correlation between genetic relationship and geographic distribution among populations in Mantel's test. For conservation of the phellodendron trees, it would be effective to consider the findings resulted from this study with ecological traits and life histories of this species.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1373-1380
• /
• 2019

In order to increase energy efficiency, developments of the advanced metering infrastructure (AMI) in the smart grid technology have recently been actively conducted. An essential part of AMI is analyzing power consumption and forecasting consumption patterns. In this paper, we analyze the power consumption and summarized the data errors. Monthly power consumption patterns are also analyzed using the k-means clustering algorithm. Forecasting the consumption pattern by each household is difficult. Therefore, we first classify the data into 100 clusters and then predict the average of the next day as the daily average of the clusters based on the deep neural network. Using practically collected AMI data, we analyzed the data errors and could successfully conducted power forecasting based on a clustering technique.

• Journal of Korean Society of Transportation
• /
• v.31 no.4
• /
• pp.85-94
• /
• 2013

The Continuous Risk Profile (CRP) has been well known to be the most accurate and efficient among existing network screening methods. However, the classical CRP uses safety performance functions (SPFs) which require a huge investment to construct a database system. This study aims to suggest a new CRP method using average crash frequencies of homogeneous groups, instead of SPFs, as rescaling factors. Hierarchical clustering analysis is performed to classify freeway segments into homogeneous groups based on the data of AADT and number of lanes. Using the data from I-880 in California, the proposed method is compared to other several network screening methods. The results show that the proposed method decrease false positive rates while it does not produce any false negatives. The method developed in this study can be easily applied to screen freeway networks without any additional complex database systems, and contribute to the improvement of freeway safety management systems.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.419-419
• /
• 2021

수자원 관리 패러다임은 공급 위주에서 수요관리로 전환되고 있다. 가용한 수자원은 한정적이나 급속한 인구증가와 도시화로 인한 물 수요의 증가로 수요관리의 효율성이 중시되고 있기 때문이다. 기존 상수도시스템은 노후화로 가동효율이 점차 낮아지고 있으며, 인력으로 월 또는 격월로 소비자의 물 사용량을 검침해 실시간 관리가 불가능하여 수요와 공급의 불균형을 초래한다. 이러한 문제를 해결할 대안으로 IT 기술과 전통적인 물관리 기술을 접목한 Smart Water Grid는 양방향 통신장치를 이용해 실시간으로 소비자의 물 사용량을 모니터링한다. 물 사용 특성을 잘 파악하면 보다 정확한 물 수요 예측이 가능하다. 특히 소비자들의 시간별, 평일, 주말, 그리고 주별 물 사용 특성을 파악하면 미래 물 수요 예측에 도움이 된다. 예측된 물 수요량에 따라 물 공급 배분 계획을 수립하여 운영 효율성을 높일 수 있다. 물 수요예측 방법 중 k-mean 군집분석은 시간별 물 사용량을 이용해 서로 유사한 여러 개의 부분집합으로 할당하여 분류하는 Machine learing 방법으로 물 사용의 유사성을 파악할 수 있다. SWG 연구단은 2019년 Vietnam Hai Duong province에 SWG Pilot plant를 구축하고 27개의 Smart water meter를 설치하여 운영하고 있다. 이에 본 연구에서는 소비자의 물 사용 특성을 분석하기 위해 27개 SWM로부터 수신된 2019년 11월 14일부터 2020년 12월 3일까지 1시간 단위의 물 사용량 데이터를 수집하였다. 그리고 k-mean 군집 방법을 이용해 시간별, 평일, 주말, 그리고 주별 물 사용 특성을 분석하였다. 이 때 최적의 군집 개수 결정을 위해 Elbow 방법을 적용하였다. 분석 결과 각 소비자의 물 사용량 특성에 따라 평균 물 수요패턴 추정이 가능하며, 향후 물 수요 예측에 도움이 될 것으로 사료된다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.291-293
• /
• 2012

SNS, 블로그의 이용이 늘어나면서, 문서의 복제와 재생산이 빈번하게 발생함에 따라 대용량 문서에서의 유사중복문서 검출이 큰 이슈로 제기되고 있다. 본 논문에서는 한국어 문서를 대상으로 이러한 문제를 해결하기 위해 품질을 유지하면서 신속하게 문서집합 중 유사중복문서를 검출하는 방법에 대해 제안한다. 제안하는 알고리즘에서는 문서를 대표하는 고빈도 Unigram Token을 활용하여 문서를 군집화함으로써 비교 대상을 최소화 하였다. 실험결과, 76만 문서에서 기존 방법 대비 평균 0.88의 Recall을 유지하면서도 중복을 검출하는데 있어서 십수초내에 처리가 가능함을 보였다. 향후 대용량 검색시스템 및 대용량 이미지, 동영상 유사중복 검출에도 활용할 수 있을 것으로 기대한다.