基于独立分量分析的遥感影像非监督分类方法
Method for Unsupervised Remote Imagery Classification Based on Independent Component Analysis
-
摘要: 利用独立分量分析的方法,从图像信号分离的角度出发,将每个波段像元的光谱特征看成是由相互独立的不同地物类型光谱信号混合而成。通过ETM+ 遥感影像数据的分类试验,验证了该方法应用于多光谱遥感影像非监督分类的有效性Abstract: This paper introduces a linear spectral random mixture analysis model based on ICA. It assumes that the spectral signature of an image pixel is linearly mixed by the spectral signature of independent materials. Experimental results with real imagery data show that the method is effective in multi\|spectral remote sensing image classification.
下载:
