大气参数查找是遥感图像大气校正中一个重要的步骤。本文针对当前多数大气参数查找表普适性不强、查找表数据存储空间巨大、具体实施步骤不详细的问题，建立了一个通用化的多维大气参数查找表，其光谱分辨率为1 nm，并适用于多种遥感传感器。本文采用二进制文件方式存储查找表，并研究采用多维拉格朗日和多维反距离加权插值方法对输入参数进行插值。结果表明以二进制文件方式存储查找表，不但可以最大化地减少磁盘存储空间，还可以实现查找表数据快速随机读取;相对于多维反距离加权插值方法，多维拉格朗日插值方法的速度更快，精度更高，与MODTRAN4.0模型计算结果（真值）相比，它的误差仅为±1.0%左右。

Atmospheric parameters are crucial to the atmospheric correction of remote sensing images. However, the universality of most atmospheric parameters Look-Up Table (LUT) at present is not strong enough, e.g., the LUT data storage space is huge and the specific implementation steps are not clear. We build a universal multidimensional atmospheric parameters LUT which not only has the 1 nm spectral resolution but also is suitable for multiple remote sensing sensors. The LUT is stored as binary data and adopted the multidimensional Lagrange and multidimensional inverse distance interpolation method to input parameters. The results of the study show that with binary data stored look-up table can both reduce disk storage space to the hilt and achieve fast look-up table data random read. Relative to the multidimensional inverse distance interpolation method, multidimensional Lagrange interpolation method is faster, of higher precision with error about 1.0%.