【作者】 刘咏梅； 杨勤科； 汤国安； 温仲明；

【Author】 LIU Yong-mei, YANG Qin-Ke, Tang Guo-an, WEN Zhong-ming(l.Department of Urban and resource Sciences, Northwest University, Xi’an, Shaanxi 710069; 2.Institute of Soil and Water Conservation, Chinese Academy of Sciences and Ministry of Water Resources, Yangling 712100)

【机构】 西北大学城市与资源学系； 中国科学院；

【摘要】 陡坡开垦使坡耕地面积扩大是造成黄土高原水土流失严重、生态恶化的根本原因。应用遥感技术,及时准确地掌握坡耕地的分布特征及面积数据,对合理实施退耕还林还草工程意义重大。本文以陕北黄土丘陵沟壑区为研究区域,以TM图像为主要信息源,采用非监督分类与监督分类相结合的混合分类方法提取坡耕地信息。通过改进采样方法,在非监督分类生成的初始训练样本的基础上,进行删除、增补、合并等样本调整,使训练样本的选取精度大大提高,明显提高了分类精度。研究证明,在黄土丘陵沟壑区,以混合分类方法提取坡耕地取得了良好的分类效果,是进行坡耕地遥感调查较为理想的方法。更多还原

【Abstract】 Reclaiming in steep slope land to extend the area of slope field plays a very important role in soil and water loss and eco-environment deterioration in loess plateau. In order to convert farmland into forest in slope land reasonably, obtaining up-to-date and reliable information on spatial distribute and magnitude of slope field by remote sensing is of critical significance. Taking loess hill and gully area of northern Shaanxi Province as a test area and Landsat TM5 data as data source, applying the integration of supervised classification and unsupervised classification, slope field and other categories are extracted. By improving the signatures selecting accuracy, this method improves classification accuracy greatly. The result shows that the integration classification is suitable to slope field investigation in the loess hill and gully area.更多还原