摘要
为更准确提取农牧交错地带的土地利用信息,本文以靖边县为例,利用分层分类法对该区土地利用类型进行分类。在TM图像光谱特征分析的基础上,提出分层分类的分类步骤,并利用ERDAS建模,逐层提取土地类型,将该区的土地利用类型分为9类。结果表明,与传统的监督分类法比较,分层分类法的分类精度有了显著的提高,对农牧交错地带的土地利用分类结果较为理想,总分类精度达到87.41%,Kappa系数也达到了85.53%。为今后该地区的土地利用信息的提取提供了有效的方法。
Taking agriculture and pasturage interlaced region of Jingbian County in Shaanxi Province as a study area,the region was classified into nine categories using the hierarchical classification method.Based on the analysis of spectral characteristics of TM images,landuse categories were obtained layer by layer using model maker of ERDAS software.The results showed that,contrasted with the traditional supervised classification method,the accuracy of hierarchical classification method was significantly improved,and the overall classification accuracy was 87.41%,while Kappa coefficient reached 85.53%.
出处
《干旱地区农业研究》
CSCD
北大核心
2010年第5期237-240,246,共5页
Agricultural Research in the Arid Areas
基金
国家自然科学基金项目(30872073)
关键词
分层分类
土地利用
光谱特征
靖边县
hierarchical classification
land use
spectral characteristics
Jingbian County
