基于Sentinel-2卫星数据的内陆湖泊水体大气校正

利用Acolite、FLAASH和C2RCC 3种大气校正算法，参考内陆湖泊历史实测遥感反射率数据，开展针对Sentinel-2卫星数据的内陆湖泊水体大气校正研究。试验结果表明：在数值方面，C2RCC校正结果和实测结果位于相同数量级，Acolite和FLAASH校正结果高估；在光谱曲线变化趋势方面，C2RCC校正结果与实测数据相近，Acolite和FLAASH校正结果存在异常，因而推荐使用C2RCC大气校正Sentinel-2的内陆水体数据；如果只考虑前5个波段，Acolite校正结果的曲线趋势与实测光谱接近，由于校正结果高估，推荐使用波段比值以消除数值高估。

Atmospheric correction of inland lake water based on Sentinel-2 satellite data

Referring to the historical measured remote sensing reflectance data, the atmospheric correction of inland lake water acquired by Sentinel-2 satellite is carried out by correction algorithms Acolite, FLAASH and C2RCC. The experimental results show that, numerically, the correction results of C2RCC are in the same order of magnitude as the measured results, and the correction results of Acolite and FLAASH are overestimated. In terms of the variation trend of spectral curve, the correction results of C2RCC are similar to the measured data, and the correction results of Acolite and FLAASH are abnormal. Therefore, it is recommended to use C2RCC atmospheric correction for Sentinel-2 inland water data. If only the first five wavebands are considered, the curve trend of the correction result of Acolite is close to the measured spectrum. As the correction result is overestimated, it is recommended to use the band ratio to eliminate the overestimation.