• 제목/요약/키워드: 김대성

Search Result 383, Processing Time 0.031 seconds

The Hyperspectral Image Classification with the Unsupervised SAM (무감독 SAM 기법을 이용한 하이퍼스펙트럴 영상 분류)

  • 김대성;김진곤;변영기;김용일
    • Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
    • /
    • 2004.04a
    • /
    • pp.159-164
    • /
    • 2004
  • SAM(Spectral Angle Mapper) is the method using the similarly of the angle between pairs of signatures instead of the spectral distance(MDC, MLC etc.) for classification or clustering. In this paper, we applied unsupervised techniques(Unsupervised SAM and ISODATA) to the Hyperspectral Image(Hyperion) which has innumerable, narrow and contiguous spectral bands and Multispectral Image(ETM$\^$+/) for the clustering of signatures. The overall measured accuracies of the USAM and ISODATA of multispectral image were 76.52%, 53.91% and the USAM and ISODATA of hyperspectral image were 63.04%, 53.91%. From the results of our test, we report that the Unsupervised SAM is better classfication technique than ISODATA. Also we believe that the "Spectral Angle" can potentially be one of the most accurate classifier not only multispectral images but hyperspectral images.

  • PDF

Evaluation of Transfer Services based on Transit Smart Card Data (스마트카드 데이터를 활용한 역사별 연계 환승시간 서비스 평가)

  • Choi, Myoung-Hun;Eom, Jin-Ki;Lee, Jun;Kim, Dae-Sung
    • Proceedings of the KSR Conference
    • /
    • 2011.10a
    • /
    • pp.1699-1706
    • /
    • 2011
  • This study analyzed the level of service on passenger transfer between metro and bus based on transit smart card data obtained in 2010. In order to evaluate the level of service on transfer, we defined the service level specially on transfer time at metro stations. The data of passenger transfer time were used in cluster analysis to classify the service level from A to F. The results show that the average transfer time from metro to bus was 6.45 minutes. The number of stations with level of service A(approximately less than 7 minutes) and B(less than 16minutes) were found to be 215 and 227stations respectively. Also, the number of stations with the level of service C and D (greater than 20 minutes for transfer) were found to be 6 stations where any type of improvement on transfer facilities is required.

  • PDF

Analysis of Passenger Transfer Patterns Based on Transit Smart Card Data in Seoul (서울시 대중교통 통행자 환승패턴 분석)

  • Song, Ji-Young;Eom, Jin-Ki;Park, Jong-Hoon;Kim, Dae-Sung;Choi, Myoung-Hun
    • Proceedings of the KSR Conference
    • /
    • 2011.10a
    • /
    • pp.563-570
    • /
    • 2011
  • This study analyze the transfer patterns of passengers in Seoul based on transit smart card data that was observed in 2010. The smart card records maximum four times of transfer and reports that approximately 90% of trips were less than one transfer and the remains were more than 2 transfers. We focus on trips with more than 3 transfers to figure out the relationship between transit service and regional connectivities. The results show that the average travel time, distance, fare are 45 minutes, 18.3km, and 1,119(KW) respectively. We develope a map for investigating transfer patterns at a regional level(dong and gu). By doing this, three types of transfers are observed as: 1) trips of which origin and destination is either same or near, 2) trips with short distance, and 3) long distance trip with low transit connectivities.

  • PDF

A Case of Sporadic Ullrich Congenital Muscular Dystrophy Caused by a COL6A1 Mutation (COL6A1 돌연변이에 의해 발생한 산발성 Ullrich 병 1례)

  • Park, Young-Eun;Kim, Tae-Hyoung;Kim, Hyang-Suk;Kim, Dae-Seong
    • Annals of Clinical Neurophysiology
    • /
    • v.12 no.1
    • /
    • pp.27-31
    • /
    • 2010
  • Ullrich disease is a rare congenital muscular dystrophy, which is clinically characterized by generalized muscular weakness, distal joint hyperextensibility, proximal joint contractures, protuberant calcanei and high-arched palate. The disease is caused by collagen VI deficiency in interstitum and/or sarcolemma of skeletal muscles, for which mutations either in COL6A1, COL6A2 or COL6A3 are responsible. We report a girl who presented with symptoms typical of Ullrich disease, in whom the diagnosis was confirmed by immunohistochemistry and molecular genetic study.

Heat Transfer Characteristics of a Slot Jet and Circular Jets Impinging on a Flat Surface (벽면에 충돌하는 슬롯형 제트와 원형 제트의 열전달 효과 비교)

  • Kim, Hui-Hyun;Kim, Dae-Seong;Yoon, Soon-Hyun
    • Proceedings of the KSME Conference
    • /
    • 2001.06d
    • /
    • pp.540-545
    • /
    • 2001
  • An experimental study was conducted to compare the heat transfer characteristics of an impinging slot jet and three kinds of impinging circular jets. Thermochromic liquid crystal with an image processing system was employed to measure the temperature of impinging wall where constant heat flux condition was applied. The distribution of convective heat transfer coefficients were then evaluated for eight nozzle-to-surface distance settings for each jet cases. The cooling effect was linearly proportional to the number of nozzles for circular jet cases at the same nozzle exit speed. However, the heat transfer under constant volume flow rate was the most at single circular jet. It was concluded that the overall convective heat transfer was better at the circular jets than the slot jet.

  • PDF

Aggregation of Hyperion Spectral Band Using Landsat-7 ETM+ Spectral Characteristic - NDVI Application (Landsat-7 ETM+ 센서 분광특성을 이용한 Hyperion 영상의 밴드 조합 - NDVI 적용을 중심으로)

  • Kim, Dae-Sung;Kim, Yong-Il;Yu, Ki-Yun
    • 한국공간정보시스템학회:학술대회논문집
    • /
    • 2005.05a
    • /
    • pp.339-344
    • /
    • 2005
  • 하이퍼스펙트럴 데이터의 효과적인 분석을 위해 밴드 추출(Feature Extraction)이나 밴드선택(Feature Selection)에 대한 연구가 최근 많이 이루어지고 있다. 본 연구는 상대적으로 많은 밴드를 가지는 하이퍼스펙트럴 영상을 식생지수(Vegetation Index)와 같은 특수한 목적에 적용하기 위해 같은 파장대의 밴드를 조합(Band Aggregation)하여 Landsat ETM+ 영상 밴드와 동일한 영상 생성을 목적으로 한다. 이를 위해 NASA에서 제공하는 밴드별 분광특성 자료를 이용하여 밴드 조합을 위한 가중치 계산식에 적용하였으며, 밴드 선택을 위한 유효 파장대를 추출해 보았다 데이터 간 편차를 줄이기 위해 실제 1분 간격으로 촬영된 동일지역의 Hyperion과 ETM+ 영상을 사용하여 알고리즘에 적용하였고, 그 결과를 영상 간 상관계수와 NDVI 영상을 이용하여 비교 분석하였다.

  • PDF

Multi-Dimension Visualization Proposition and Clustering of Remote Sensing Data Using Star Coordinates Technique (Star Coordinates 기법을 이용한 원격탐사 데이터의 다차원 시각화 제안 및 클러스터링)

  • Kim, Dae-Sung;Kim, Yong-Il;Yu, Ki-Yun
    • 한국공간정보시스템학회:학술대회논문집
    • /
    • 2005.05a
    • /
    • pp.313-318
    • /
    • 2005
  • 단 밴드 영상과는 달리 다차원 데이터는 분광적인 특성을 이용한 자동화된 영상 분석을 수행하는 장점이 있는 반면, 3차원 이상의 데이터를 분광차익 상에 시각화 하는데 어려움이 따른다. 클러스터링 기법을 이용한 영상 정보 추출은 자동화된 영상 분석적인 측면에서 중요한 분야 중 하나로서, 분광차원에서 구 형태의 조밀한 클러스터를 분리하는데 효과적인 방법으로 알려져 있지만 부정형(不定形)의 클러스터를 추출하는 방법에는 한계를 가진다. 따라서 본 연구는 모든 차원의 데이터를 2차원 상에 시각화하여 화소간 인접성을 개략적으로 확인할 수 있는 Star Coordinates 기법을 제안한다. 데이터의 다차원 시각화를 통해, 부정형 클러스터를 제거하여 다음 단계의 영상 분석 시 발생할 수 있는 오류를 방지할 수 있고, 명확한 클러스터를 확인 지정하여 클러스터링 정확도를 골일 수 있을 것으로 기대된다. 부가적인 연구고서, Star Coordinates 기법을 적용하여 Plot된 영상 데이터를 K-Means 알고리즘을 이용한 무감독 분류를 수행하여 그 결과를 확인하였다.

  • PDF