卫星遥感数据分类不确定性的容差粗糙集处理

Tolerant Rough Set Processing on Uncertainty of Satellite Remote Sensing Data Classification

提出一种基于光谱特征邻域的容差粗糙集分类方法,用来处理卫星遥感数据分类中的不确定性问题。利用北京地区Landsat-5TM数据进行分类试验,将结果与现有容差粗糙集方法分类结果进行对比和统计。结果表明,该方法具有良好的可理解性和稳定性,能够处理卫星遥感数据分类的不确定性,在具有复杂光谱特征地物分类方面具有发展潜力。

Based on the spectral feature neighborhood, this paper proposes a tolerant rough set classification method to handle the uncertainty in the process of satellite remote sensing data classification. Experiment is carried out with Landsat-5 TM image of eastern Beijing. Classification result is compared with result from the current tolerant rough set method. Outcome indicates that the method is more interpretable and reliable, and effectively handles the uncertainty in the process of satellite remote sensing classification. It is a promising tool at classifying areas with complex spectral feature.