为了充分利用中等分辨率遥感影像的空间、纹理和光谱信息,解决影像中存在的大量混合像元问题,提出了基于地理分区的多端元光谱混合分析方法。实验结果表明,基于地理分区的多端元光谱混合分析方法消除了多端元光谱混合分析方法造成的像元过度分解现象,能够得到比较连续且和实际地物分布较为吻合的分析结果,其分解精度和效率优于多端元光谱混合分析方法。 In order to make full use of the spatial, texture and spectral information of medium resolution remote sensing image and solve the problem of large number of mixed pixels in the image, a multi-terminal spectral hybrid analysis method based on geographic partitioning is proposed. The experimental results show that the multi-component spectral hybrid analysis method based on geographic partitioning eliminates the phenomenon of pixel over-decomposition caused by the multi-component spectral hybrid analysis method and can obtain the analysis results that are more consistent and consistent with the distribution of physical features. The decomposition accuracy and Efficiency is better than multi-end element spectrum hybrid analysis method.