根据MERIS(Medium Resolution Imaging Spectrometer,中等分辨率成像光谱仪)L1B产品特点,以台湾海峡为研究区域,开展MERIS数据大气校正研究。在阐述水色遥感数据大气校正原理的基础上,首先进行水体像元提取和辅助参数空间内插,获取大气校正所需的数据集。接着进行耀斑反射率计算和校正并利用加倍法精确计算瑞利散射。最后针对台湾海峡水体为清洁到轻度浑浊,提出先区分水体类别,再分别利用epsilon指数外推法进行一类水体气溶胶散射计算,和利用神经网络法计算二类水体离水反射率即阢的方法。对大气校正的结果进行ρw负值像元数统计和光谱分析表明,该方法不仅降低阳为负值的像元数,而且使大气校正后水体的光谱曲线同现场数据保持较好的一致性。
Taking the Taiwan Straits as a test site, the study of atmospheric correction was carried out on MERIS (Medium Resolution Imaging Spectrometer)data according to the characteristics of MERIS LIB product. The principles of atmospheric correction were introduced in the first part of this paper. The first step of the study is to obtain the datasets for atmospheric correction by extracting water body pixels from the image and interpolating the auxiliary parameters. The glint reflectance was calculated and corrected, and exact Rayleigh scattering was calculated by adding-doubling method subsequently. Finally, based on the fact that water in the Taiwan Straits is from being clear to slightly turbid, a method was proposed to calculate aerosol scattering of case Ⅰ water by using an epsilon exponent extrapolation algorithm and water-leav ing reflectance of case Ⅱ water by a neural network algorithm accordingly, following water type discrimina tion. With analyzing the statistics of negative tow pixel numbers and spectral signatures of the result, the conclusion can be drawn that the method proposed in this paper reduces the number of negative ρw pixels while keeping the spectral signatures of water body consistent with those of in situ data well.