天文学家拍摄的天体光纤光谱被CCD记录下来之后,得到的仅仅是二维光谱图像。而这些光谱图像无法直接使用,需要借助一定的抽谱方法把二维CCD光谱图像转换成一维光谱。抽谱方法大致分为四类:孔径抽谱方法、优化的孔径方法、轮廓拟合方法和反卷积方法。本文将详细介绍这些方法的优缺点和适用范围。同时结合郭守敬望远镜(LAMOST)CCD光谱图像的特点对这些方法进行讨论。 Celestial fiber spectra taken by astronomers are recorded by the CCD, the result is only two-dimensional spectral images. However, these spectral images can not be used directly, and it is necessary to convert the two-dimensional CCD spectral image into a one-dimensional spectrum by using a certain method of spectrum extraction. There are four broad categories of spectral methods: Aperture Radar, Optimized Aperture, Contour Fitting and Deconvolution. This article will detail the advantages and disadvantages of these methods and their applicability. At the same time, these methods are discussed based on the characteristics of LAMOST CCD spectral image.