1. NCCU Academic Hub
  2. 學術產出
  3. College of Social Science
  4. Proceedings
  5. Item

學術產出-Proceedings

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

  • No doi shows Citation Infomation
題名 Using object-based classification to detect landslides sites using high resolution aerial images
作者 Kung, Fan-En;Jan, Jan J.-F.;Shao, Y.-C.;Li, M.-Y.;Yeh, K.-S.;Chen, L.-H.
孔繁恩
貢獻者 地政學系
關鍵詞 Aerial images; Classification approach; Classification methods; High-resolution aerial images; Hybrid classification; Object-based classifications; Segmentation techniques; Spatial resolution; Image classification; Landslides; Pixels; Remote sensing; Image segmentation
日期 2012
上傳時間 10-Apr-2015 16:34:39 (UTC+8)
摘要 A variety of image classification methods have been applied in satellite images analysis, including supervised, unsupervised, and hybrid classifications. However, many researches indicated that traditional pixel-based classification approaches often resulted in less satisfactory outcome when applied to high resolution aerial image data. The main reason is that pixel-based approaches often raise over-classification problem as the spatial resolution of images increases. The objective of this study is to use object-based classification method to detect landslides sites in aerial images acquired by Z/I DMC and Leica ADS40. Firstly, multi resolution segmentation technique will be applied to segment image into regions that correspond to various areas of interest. Then each region will be classified into appropriate land cover types by using different kinds of indexes which are defined in this study, and the landslide sites will be identified. Finally, the result of landslides sites can show that the object-based classification is a good method for extracting the areas of landslides.
關聯 33rd Asian Conference on Remote Sensing 2012, ACRS 2012
資料類型 conference
dc.contributor 地政學系
dc.creator (作者) Kung, Fan-En;Jan, Jan J.-F.;Shao, Y.-C.;Li, M.-Y.;Yeh, K.-S.;Chen, L.-H.
dc.creator (作者) 孔繁恩zh_TW
dc.date (日期) 2012
dc.date.accessioned 10-Apr-2015 16:34:39 (UTC+8)-
dc.date.available 10-Apr-2015 16:34:39 (UTC+8)-
dc.date.issued (上傳時間) 10-Apr-2015 16:34:39 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/74472-
dc.description.abstract (摘要) A variety of image classification methods have been applied in satellite images analysis, including supervised, unsupervised, and hybrid classifications. However, many researches indicated that traditional pixel-based classification approaches often resulted in less satisfactory outcome when applied to high resolution aerial image data. The main reason is that pixel-based approaches often raise over-classification problem as the spatial resolution of images increases. The objective of this study is to use object-based classification method to detect landslides sites in aerial images acquired by Z/I DMC and Leica ADS40. Firstly, multi resolution segmentation technique will be applied to segment image into regions that correspond to various areas of interest. Then each region will be classified into appropriate land cover types by using different kinds of indexes which are defined in this study, and the landslide sites will be identified. Finally, the result of landslides sites can show that the object-based classification is a good method for extracting the areas of landslides.
dc.format.extent 176 bytes-
dc.format.mimetype text/html-
dc.relation (關聯) 33rd Asian Conference on Remote Sensing 2012, ACRS 2012
dc.subject (關鍵詞) Aerial images; Classification approach; Classification methods; High-resolution aerial images; Hybrid classification; Object-based classifications; Segmentation techniques; Spatial resolution; Image classification; Landslides; Pixels; Remote sensing; Image segmentation
dc.title (題名) Using object-based classification to detect landslides sites using high resolution aerial images
dc.type (資料類型) conferenceen