基于热红外像元分解的裸土信息自动提取方法

基于热红外波段像元的分解与重构,对缨帽变换指数和归一化裸土指数进行逻辑组合,提出了一种新的裸土指数Bareness Index,并利用Landsat ETM+数据对珠江三角洲进行了裸土信息自动提取实验。结果表明:该方法在增强裸土信息的同时,可以有效抑制背景地物的干扰,并提高了检测精度,信息提取精度为84.7%。研究对区域土地利用类型分类精度的提高具有一定的理论与现实意义。

Automatic Method of Bare|soil Areas Extraction based on Pixel Decomposition

Based on decomposing and composing the thermal infrared band pixel,with the logical combination of the tasseled cap transformation (K-T) and Normalized Difference Bareness Index (NDBaI),a new Bareness Index (BI) was developed.And it has been applied in extracting the bare|soil areas in Pearl River Delta using Landsat ETM+ data in 2003.The results show that the BI not only has a good effect on the enhancement of bare soil information,and the inhibition of the background information,but also improve the accuracy of detection.The results of this study could be of scientific and practical merits in regional remote sensing monitoring and improve the accuracy of land use classification.