ランドサットTMデータによる都市内緑地の抽出

DOI Web Site 被引用文献2件

書誌事項

タイトル別名
  • Classification of Urban Green Spaces Using Landsat TM Data
  • ランドサット TM データ ニ ヨル トシ ナイ リョクチ ノ チュウシュツ

この論文をさがす

抄録

Landsat TM data is analyzed to give the distribution of green spaces in urban area with the three methods of classification; i. e. maximum likelihood decision rule, image ratioing technique and fuzzy clustering. The possibility of detecting small green spaces whose scales are less than a pixel is considered. The resulted distribution is compared with the ground truth to evaluate the efficiency of the methods.<BR>Results show that maximum likelihood decision rule is not suitable for the analysis of urban cover which consists of various types of surfaces. Result of the ratioing technique generally agrees with the ground truth in spite of the simplicity of the method.<BR>With fuzzy clustering, the distribution of green spaces agrees well with the ground truth and it can be said that small green spaces are well detected. This method is the most effective method in classifying urban surfaces among these tree methods.

収録刊行物

  • 造園雑誌

    造園雑誌 50 (5), 305-309, 1986

    社団法人 日本造園学会

被引用文献 (2)*注記

もっと見る

詳細情報 詳細情報について

問題の指摘

ページトップへ