%0 Journal Article
%T The Information Extraction of Land Use in the Composite Zone of High-Intensity Artificial Disturbance Based on Multivariate Methods Coupling<br>基于多元方法耦合的高强度人为扰动复合区土地利用信息提取
%A WANG Jinliang
%A SHAO Jing''an
%A LI Yangbing
%A <br>王金亮
%A 邵景安
%A 李阳兵
%J 资源科学
%D 2012
%I 
%X The land surface stave caused by the high-intensity artificial disturbance makes the phenomena of"same objects with different spectrum"and"different objects with the same spectrum"especially serious. Depending on the spectrum characteristics of surface features alone to extract land utilization information has obvious short points. Therefore, it is very necessary to seek an extraction model which is fit in the condition of a specific region and is based on multivariate methods coupling. In this paper, we first found the spectrum characteristics of typical land use pattern and the changing law of normalized difference indexes, and then coupled the spectral information and established the decision function with the interactive date language (IDL). TM images of 1993, 2001 and 2009 were classified with decision tree so as to automatically extract the land utilization change information in the composite zone of backwater submergence and rapid urbanization in the Three-Gorges Reservoir Area. The results indicate that: 1) The distribution patterns of spectral response curve trend and normalized difference exponential curve of typical land pattern accord with the three fundamental land type features of ground surface, which proves the rationality of the three indexes and the validity of the selection in each sample area; 2) Land use pattern is difficult to be extracted on TM band 1, 2 and 3 because of the nonobvious differences and strong relativity of the change trend of spectral characteristics. As for TM band 3, 4, and 5, it is easy to extract the pattern because of the great change on spectral characteristics and the rich land use information; 3) The value change and curve distribution pattern of the three indexes and their interpolation combination indexes are in line with the actual change of land use in study area; 4) The accuracy of decision tree classification with multivariate methods was increased by 18.46%, 15.43% and 24.18% in 1993, 2001 and 2009 respectively compared with that of the classic automatic classification technology. Accordingly, the decision tree method based on multivariate methods coupling is effective for land use change information ectraction in the composite zone of artificial disturbance in Three-Gorges Reservoir Area, and it can provide an integration concept of the existing methods for the information extraction.
%K Land use
%K Information extraction
%K Multivariate methods coupling
%K Composite zone of high-intensity artificial disturbance
%K The Three-Gorges Reservoir Area<br>土地利用
%K 信息提取
%K 多元方法耦合
%K 高强度人为扰动复合区
%K 三峡库区
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=B5EDD921F3D863E289B22F36E70174A7007B5F5E43D63598017D41BB67247657&cid=B47B31F6349F979B&jid=9DEEAF23637E6E9539AD99BE6ABAB2B3&aid=ADCA14AF4CF99CB26E34AA2753A14FF4&yid=99E9153A83D4CB11&vid=339D79302DF62549&iid=E158A972A605785F&sid=583C993D4E08F5CE&eid=CFBDB06850C21CC6&journal_id=1007-7588&journal_name=资源科学&referenced_num=0&reference_num=25