%0 Journal Article
%T Study on Water Bodies Extraction and Classification from SPOT Image<br>SPOT卫星影像的水体提取方法及分类研究
%A DU Jin-kang
%A HUANG Yong-sheng
%A FENG Xue-zhi
%A WANG Zhou-long
%A <br>都金康
%A 黄永胜
%A 冯学智
%A 王周龙
%J 遥感学报
%D 2001
%I 
%X The remote sensing mechanism and spectral characteristics of water bodies and other objects in SPOT 4 imagery were analyzed. Due to the similar infrared spectral response feature of water bodies to those of the objects under the mountain shadows in the imagery, thresholding for infrared bands is not suitable for the extraction of water bodies. The decision tree algorithm including several classifiers was put forward, which can extract water bodies effectively and easily. The terrain information and geometric features such as perimeter, area, and shape, which are frequently used in visual interpretation, can be used to classify water bodies into subclass. The decision tree algorithm including several classifiers associated with terrain and geometric information was developed to classify water bodies. Jiangning county of Jiangsu Province was selected as a case study area, SPOT 4 (XI, 1999.1) and contour data of the area were used as information sources, the results after applying the proposed approach were satisfactong.
%K water bodies
%K decision tree algorithm
%K spatial structure<br>水体
%K 决策树分类
%K 空间结构
%K SPOT卫星影像
%K 提取方法
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=E62459D214FD64A3C8082E4ED1ABABED5711027BBBDDD35B&cid=A41A70F4AB56AB1B&jid=F926358B31AC94511E4382C083F7683C&aid=C602CDA544BDD00F&yid=14E7EF987E4155E6&vid=94C357A881DFC066&iid=38B194292C032A66&sid=797D49279EA93BC4&eid=D5C73DEF4CF8FAF3&journal_id=1007-4619&journal_name=遥感学报&referenced_num=87&reference_num=14