中文摘要：

对火星高光谱遥感数据进行混合像元分解,有助于获取像元内部火星表面矿物含量。端元光谱提取和光谱解混是混合像元分解的关键技术。以ORB0942_2轨道覆盖的火星南极地区作为研究区,应用纯净像元指数法（PPI）从影像中提取出端元光谱,并利用线性分解模型对影像中混合像元进行了分解,计算出其各端元组分的百分含量,获得了研究区水冰、石膏、钙镁橄榄石及紫苏辉石的相对含量分布图。

英文摘要：

Spectral unmixing can contribute to the mapping of Mars minerals with hyperspectral remote sensing image.End member spectra extraction and unmixing algorithm are the two key factors for pixel unmixing.In this paper,one image in Antarctica of Mars（ORB0942_2） is selected as the test image to map the contents of minerals using spectral unmixing.The pixel purity index（PPI） method is used to extract the end member spectrum from the image,and the linear unmixing model is applied to decompose the image of mixed-pixel.The percentage of each end member are calculated.The distribution of H2O-ice,gypsum,monticellite and hypersthene are mapped respectively.