多维空间角度方法提取植被覆盖区蚀变信息

基于光谱角制图法，提出运用多维空间角度方法来提取植被覆盖较好区域的遥感图像蚀变信息。采用目视解译方法得到的样本向量代替从波谱库中得到的端元组分矢量来计算多维向量间的空间角，并且加入合适的约束条件，来改进多维空间角度方法。给出了植被覆盖区蚀变信息提取的实施方法与步骤。最后运用改进的多维空间角度方法提取实际区域的蚀变信息，并将它与用比值法和主成分分析法得到的结果作比较。表明改进的多维空间角度方法提取蚀变信息的效果比较理想，与该地区的地质图件更为一致，从而证实了该方法的有效性。

Alteration Information Abstraction at Vegetation Covered Area Using Multidimensional Space Angle Method

ion from vegetation covered area has been presented in the light of spectral angle mapping in this paper. Through computing the multidimensional space angle by substituting the sample vectors from visual interpretation for the endmember vectors from spectral library and introducing suitable constraint conditions, the multidimensional space angle method is improved greatly. Also the practical implementation procedure adapted to alteration information abstraction is explored. Finally, we use the improved method to abstract the alteration information of a concrete study area and compare the results with those obtained from the band ratio method and the principal component analysis method. The results show that the effects of abstracted alteration information using multidimensional space angle method are more satisfying and better correspond to the actual geological status, therefore the validity of this method has been proved.