中文摘要：

环境卫星（HJ-1A/B）的CCD图像数据在内陆水环境遥感监测中具有很大的潜力,但它的大气校正的精度是影响其定量化应用的重要因素。文章以上海淀山湖地区HJ-1A卫星搭载的CCD1多光谱数据为应用实例,首先实现了瑞利散射的精确计算,在气溶胶散射的计算中,由于内陆湖泊Ⅱ类水体的光学特性,致使环境星CCD数据的近红外波段的离水辐射不为零,因此文章借助同步MODIS数据的短波红外波段获取了大气校正因子,并利用清洁水体像元在绿光波段气溶胶散射较稳定的特点,分离出离水辐射量和气溶胶的贡献,实现了CCD数据各波段的大气校正,反演出离水辐射亮度,并通过水气界面转换,获得了水体的遥感反射率。通过同步的实测数据对比表明,该文的大气校正算法可以有效地对环境小卫星多光谱数据进行大气校正。

英文摘要：

The CCD multi-band data of HJ-1A has great potential in inland water quality monitoring,but the precision of atmospheric correction is a premise and necessary procedure for its application.In this paper,a method based on dark pixel for water-leaving radiance retrieving is proposed.Beside the Rayleigh scattering,the aerosol scattering is important to atmospheric correction,the water quality of inland lakes always are case Ⅱ water and the value of water leaving radiance is not zero.So the synchronous MODIS shortwave infrared data was used to obtain the aerosol parameters,and in virtue of the characteristic that aerosol scattering is relative stabilized in 560 nm,the water-leaving radiance for each visible and near infrared band were retrieved and normalized,accordingly the remotely sensed reflectance of water was computed.The results show that the atmospheric correction method based on the imagery itself is more effective for the retrieval of water parameters for HJ-1A CCD data.