基于光谱理论的遥感信息提取方法的研究  

Research on Remote Sensing Information Extraction Method Based on Spectral Theory

作　　者：邹忠睿[1] 杨为民[1] 牛远[2] 姜岩[2] 魏锦宏[1] 

机构地区：[1]西南林业大学林学院,昆明650224 [2]中国环境科学研究院,北京100012

出　　处：《现代计算机》2014年第3期15-18,27,共5页Modern Computer

摘　　要：以东昌府区为例,根据区内各种地物不同的光谱特征,利用遥感光谱分析方法,基于ERDAS IMAGINE平台对东昌府区内各种地物分别用监督分类和非监督分类两种方法进行信息提取,对两种方法进行比较,选择效果比较好的一种对东昌府区遥感信息进行提取研究,并对提取结果进行分析。研究表明,基于光谱理论的遥感信息监督分类提取方法在对东昌府区进行信息提取时更为精确。Takes Dongchangfu district as an example, according to the various surface features of different spectral characteristics. Uses remote sensing spectral analysis method, based on the software platform of ERDAS IMAGINE, uses a variety of surface features in Dongchangfu area supervised classification and unsupervised classification of two methods of information extraction, Comparing the two methods, better selection effect a kind of remote sensing information extraction Dongchangfu study area. The results show that the supervised classification of remote sensing information extraction method based on the theory of the spectrum is more accurately about in information extraction in Dongchangfu.

关 键 词：光谱分析 地物信息提取 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]