高光谱遥感与传统的单波段、多光谱数据相比，波段量大量增加、波段宽度极大降低，对地面目标的光谱特性的测度更加细致，然而波段的增多必然导致数据量急剧增加、计算量增大、信息冗余增加以及统计参数的估计偏差增大。因此，对高光谱数据进行降维处理具有重要意义。一方面，降维能够使图像远离噪声，提高图像数据质量；另一方面，能够去除图像中的无价值波段，减少波段数，从而降低计算量，提高运算效率。主成分分析是常用的高光谱数据降维处理方法之一。(Compared with the single band, hyperspectral remote sensing and traditional multi spectral data, a substantial increase in the amount of band width of band, greatly reduced, measure spectral characteristics of ground objects in greater detail, but the increase will inevitably lead to a sharp increase in band data and increase the amount of calculation, and increase the information redundancy of statistical parameter estimation error. So it has important significance to reduce the dimensionality of hyperspectral data. On the one hand, the dimension reduction can make the image far away from noise and improve the quality of image data; on the other hand, it can remove the non value bands in the image, reduce the number of bands, thereby reducing the amount of calculation and improving the efficiency of the calculation. Principal component analysis is used for dimensionality reduction of hyperspectral data processing methods.)