干旱区生态环境知识资源中心(Arid): 新疆干旱区绿洲土壤盐渍化信息提取对比研究

Arid

	新疆干旱区绿洲土壤盐渍化信息提取对比研究
其他题名	Contrast Study on Extracting of Soil Salinization Arid Area of Xinjiang Oasis
	张飞; 塔西甫拉提·特依拜; 丁建丽; 依力亚斯江·努尔麦麦提; 田源
ISSN	1004-0323
出版年	2008
卷号	23 期号:4 页码:398-404
中文摘要	在遥感影像分类的过程中非光谱特征起着重要的辅助作用。纹理特征作为一种重要的非光谱特征对于遥感影像分类精度的提高也有很重要的作用。以渭干河-库车河三角洲绿洲为例，利用ETM＋数据，探讨了该绿洲盐渍化土地覆盖信息的提取方法。提出了基于SVM的光谱和纹理两种信息复合的分类方法，通过此方法对该绿洲进行分类研究，并将分类结果与最小距离法、最大似然法（MLC）、神经网络法（Neural net）和单源数据（光谱）SVM分类结果进行定性和定量比较分析。研究结果表明：该方法能够有效地解决单数据源分类效果破碎、分类精度不高等问题，并对高维输入向量具有较高的推广能力。总精度达到93．1795％，比单源信息的SVM分类法提高了3．1618％，比最大似然法提高了4．8252％，比神经网络法提高了7．4756％，而与最小距离法相比，总精度甚至提高了11．1029％，取得了良好的效果。与传统的分类方法的比较表明，文中所提出的分类方法具有明显的优越性和良好的前景，因此该方法更适合于遥感图像分类和盐渍化信息提取，是地物遥感信息提取的有效途径。
英文摘要	The non-spectral characteristics play an important role in assisting the image classification in remote sensing,as these characteristics can avoid the mistake classification made by the ＂the same object with different spectrum＂ phenomenon. The texture characteristic is one kind non-spectral characteristic. It is also helpful to improve the classification precision. In this paper, the author takes the Delta Oasis of Weigan and Kuqa rivers for example,using ETM＋ data; discussing the method of extracting of soil salinization. This paper reports the classification method based on support vector machine（SVM）,and introduced the fundamental theory of SVM algorithm, then incorporating of spectrum and texture information. The classification result is compared with minimum distance classification, maximum likelihood classification, neural net classification and single data source（spectrum）SVM classification qualitatively and quantitatively. The research result shows that this method can effectively solve the problem of low accuracy and fracture classification result in single data source classification, it has high spread ability toward higher array input, the overall accuracy is 93. 1795 %,which increases by 3. 1618% comparing with single data source SVM and increases by 4. 8252% comparing with maximum likelihood classification, and increase by 7. 4756% comparing with neural net classification,even increases by 11. 1029% comparing wit minimum distance classification and thus acquires good effectiveness. The classification results were easier interpreted when compared with the conventional classification method. Therefore, the classification method based on SVM（Support Vector Machlne）and incorporate the spectrum and texture information can be adapted to RS image classification and monitoring of soil salinization, furthermore, provides and effectives way for the things remote sensing information extraction.
中文关键词	支持向量机（SVM） ; 盐渍化 ; 灰度共生矩阵 ; 纹理信息
英文关键词	Support vector machines（SVM） Soil salinization Gray co-occurrence matrix Texture information
语种	中文
国家	中国
收录类别	CSCD
WOS类目	REMOTE SENSING
WOS研究方向	Remote Sensing
CSCD记录号	CSCD:3381385
来源机构	新疆大学
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/222979
作者单位	新疆大学资源与环境科学学院, 乌鲁木齐, 新疆 830046, 中国
推荐引用方式 GB/T 7714	张飞,塔西甫拉提·特依拜,丁建丽,等. 新疆干旱区绿洲土壤盐渍化信息提取对比研究[J]. 新疆大学,2008,23(4):398-404.
APA	张飞,塔西甫拉提·特依拜,丁建丽,依力亚斯江·努尔麦麦提,&田源.(2008).新疆干旱区绿洲土壤盐渍化信息提取对比研究.,23(4),398-404.
MLA	张飞,et al."新疆干旱区绿洲土壤盐渍化信息提取对比研究".23.4(2008):398-404.