應用SPOT衛星影像與地理資訊於林地土地利用型綠度分析

Applying SPOT Imagery and Geographic Information in Greenness Analysis on Forest Land-use

謝漢欽(Han-Ching Hsieh)

《台灣林業科學》 11卷1期 (1996/03) Pp. 77-86

https://doi.org/10.7075/TJFS.199603.0077

摘要

本研究使用兩期的SPOT多光譜影像及一期的全色影像，針對東勢林區的大安區事業區林地使用型，進行常態化差異植生指數的綠度分析，於影像常態化處理中減低了影響直接反射輻之散射、天空光效應及地形效應。結合影像分類技術於綠度分析，將分析成果依統計分析比較變遷的差異。將綠度值轉換成林地土地使用地理資訊層的一個主題欄位有利於造林成果監測及提供林地空間覆蓋率的有利資訊。將HRV影像與全色影像，以影像轉換及合成處理，亦有利於造林地狀況之監測。應用兩期影像之光譜向量變遷分析，可暸解植生綠度的變遷。

關鍵字

常態化差異植生指數 ； 綠度分析 ； 影像常態化處理 ； 影像分類

並列摘要

The objective of this study was to apply 2 SPOT HRV images obtained on different dates to analyze the greenness values on land - use units in the Donshu National Forest. The normalized difference index was used to estimate the greenness on each SPOT HEV image. In the image preprocessing, the scattering contributions from atmosphere, skylight radiance and topographic effects were relieved by employing an image normalization procedure. To measure the changes between these 2 dates, this study combined image classification and statistical techniques to evaluate the results. The evaluated greenness value on each land-use polygon was transferred to a mean value that could be linked to the attribute table of a forest land-use thematic layer in the GIS database. This procedrue will be beneficial to forest-land monitoring and will provide useful information for futher spatial analysis with GIS. In addition, merging the SPOT HRV image with a panchromatic image and applying the change vector analysis in this study also proved helpful for detection of greenness change.

並列關鍵字

normalized difference index ； greenness analysis ； image normalization procedure ； image classification