| 摘要: |
| 为了提高地物识别的正确性,探讨荒漠化土地类型自动分类的有效途径,克服异物同谱和同物异谱现象。以陕西省榆林市榆阳区为例,利用Landsat 5卫星的TM数据,探讨了荒漠化土地覆盖信息的提取方法。在对不同土地覆被类型光谱特征进行系统分析的基础上,根据归一化指数(NDVI)和差值指数(TM5-TM4)及TM3、TM5、TM7波段的光谱值,提出了分层信息提取方案,利用该方案分步提取地表覆被信息。结果表明,此方法简单、实用,是地物遥感信息提取的有效途径。 |
| 关键词: TM数据 分层信息提取 光谱曲线 波段指数 荒漠化 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金项目（30571527）； 教育部科技重点项目(03157) |
|
| Study on land cover spectral character and remote sensing information extraction of desertification land |
|
|
| Abstract: |
| In order to enhance the thing recognition accuracy,discusses the effective automatic sorting way of desertification land type,and overcome the different thing with same spectrum and the same thing with different spectrum phenomenon.Based on the desertification land,this article probes into the extraction method of land cover information in Yuyang district of Yulin city in Shannxi Province by using TM data information of Landsat 5.In the light of multi-band data of TM,the normalized index(NDVI) and the difference index(TM5-TM4) are adopted.This article has also put forward the suggestion of multi-layer information extraction strategy based on TM3,TM5 and TM7 spectrum.The results show this method is simple and practical which provides an effectives way for the things remote sensing information extraction. |
| Key words: TM data,multi-layer information extraction,spectral curve,band index,desertification |
