本文为了提高地物识别的正确性,克服异物同谱和同物异谱现象,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。文章提出了基于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)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,并对高纬输入向量具有较高的推广能力,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。
In this paper, in order to overcome the different thing with same spectrum and the same thing with different spectrum phenomenon, taking the Delta Oasis of Weigan and Kuqa rivers for example, using ETM + data, the method of extracting of soil salinization is discussed. The classifieation method based on support vector machine (SVM) and combination of spectrum and texture information is proposed. The classification result is compared with minimum distance classification, maximum likelihood 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. Therefore, this method is adapted to RS image classification and monitoring of soil salinization, furthermore, provides an efficient way for remote sensing information extraction.