中文摘要：

针对高光谱遥感地物识别中存在的问题,从光谱的波形特征、运算速度、空间细节特征的光谱区分性等方面进行了算法的改进,在此基础上提出了分形信号的算法。利用CASI高光谱数据针对算法本身的性能、效率等进行了测试,对工作区的高光谱遥感影像岩性特征进行提取。针对高光谱遥感数据分形信号的初始值、迭代步长等特征进行了讨论。分形信号算法在一定程度上更细化了相似特征高光谱的可区分性,该算法应用在CASI数据的岩性特征提取,实现了基岩裸露区域地表岩性特征精确提取。

英文摘要：

The study aiming at the problems of the distinction of spectrum waveform characteristics,operation speed,spatial detail spectral features for the improvement of the algorithm in Hyperspectral remote sensing feature recognition,On the basis of this puts forward the algorithm of the fractal signal.The performance,efficiency,etc of the algorithm itself has been tested by using CASI hyperspectral data,hyperspectral remote sensing image lithologic characteristics of the study area also has been extracted.The initial value of the signal,the iteration step length and other characteristics of fractal signal of the hyperspectral remote sensing data were discarded in this study.To a certain extent,the fractal signal algorithm can refine the distinguishability of the similar characteristics of hyperspectral,and the algorithm used for feature extraction in CASI data of lithology achieves the purpose to accurately extract the surface lithology of bedrock exposed areas.