政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/69911

政大典藏 > College of Social Science > Department of Land Economics > Periodical Articles > Item 140.119/69911

Please use this identifier to cite or link to this item: https://nccur.lib.nccu.edu.tw/handle/140.119/69911

Title:	應用衛星遙測與影像分類技術估算台南市地表不透水率
Other Titles:	Application of Satellite Remote Sensing and Image Classification Techniques for Estimating the Percentage of or Impervious Surface Area in Urban Areas-A Tainan City Case Study
Authors:	孫振義 Sun, Chen-Yi
Contributors:	地政系
Keywords:	地表不透水率;衛星遙測;監督式分類;地表分類圖;福爾摩沙衛星二號 Percentage of impervious surface area ISA;Satellite remote sensing;Supervised classification;Surface classification map;FORMOSAT-2
Date:	2008.06
Issue Date:	2014-09-15 14:45:30 (UTC+8)
Abstract:	地表不透水率為評估都市水循環體系重要的指標，在都市微氣候與生態環境的評估上，亦占有舉足輕重的影響性。為了具體呈現都市各類地表分布的狀況與特定區域內之「地表不透水率(percentage of impervious surface area, ISA)」，本研究以台南市為例，應用衛星遙測與影像分類技術估算台南市ISA。本研究將地表分類成路面、建築物、植被、裸露地與水體五類，其中路面與建築物之面積占上述五類地表總面積之百分比即為ISA。本研究採用FORMOSAT-2空間解析8公尺之多光譜影像，以監督式分類法中的「最大概似法 (maximum likelihood)」進行衛星影像分類的工作，得到台南市各類地表分類圖，其分類整體精準度為73.19%，Kappa Coefficient為0.66，適合度檢定值為0.95，△ISA為-1.43%。整體來看，本研究所得到的分類結果，對於了解台南市地表分布狀況以及計算各行政區之ISA都極具有參考價值。最後，根據台南市各類地表分類圖計算出台南市各行政區ISA，其值由高而低依序如下：中西區66.43%、北區65.63%、東區58.90%、南區34.18%、安平區34.07%、安南區24.51%，而台南市全區為32.98%。本研究證實應用衛星遙測與影像分類技術可有效估算都市區域地表不透水率，並計算出台南市各行政區ISA，未來將依本研究之成果，繼續解析影響都市地表不透水率之都市計畫相關因子。 The percentage of impervious surface area (ISA) is one of the most important indexes for assessing urban water circulation. This index is also important for evaluating the urban micro climate and eco-environment. In this study, satellite remote sensing and image classification techniques were used to estimate the percentage of impervious surface area in Tainan City. This paper classifies the city land cover as roads, buildings, bare-land, vegetation or water, and ISA is defined as the percentage of roads and buildings in the five surface areas. A FORMOSAT-2 8-m resolution multi-spectral image was classified by supervised maximum likelihood classifier. The overall accuracy was 73.19%, and Kappa Coefficient was 0.66. The goodness of fit was 0.95, and △ISA was-1.43%. The classifications were used to construct a Tainan City surface classification map of urban land cover. The derived map was used to estimate ISA in all administrative areas in Tainan City. The ISA value were 66.43%; 65.63%; 58.90%; 34.18%,; 34.07% and 24.51% in the center-western, north, east, south, Anping and Annan districts, respectively, and the overall value in Tainan was 32.98%. We arrive at the conclusion that application of satellite remote sensing and image classification techniques are accurate for estimating the percentage of impervious surface area in urban areas. A future study may examine the relationship between ISA in all administrative areas and urban planning factors in Tainan City.
Relation:	都市與計劃, Vol.35, No.2, pp.123-139
Data Type:	article
Appears in Collections:	[Department of Land Economics] Periodical Articles

Files in This Item:

File	Description	Size	Format
2313.pdf		3135Kb	Adobe PDF2	902	View/Open

社群 sharing

Loading...