论文部分内容阅读
大气订正是遥感信息定量化研究中必不可少的一步,目前已有一些成熟的方法,但由于HJ-1A/B卫星CCD相机波段设置特点,常规的大气订正方法基本不适合于HJ-1A/BCCD影像.本文在大量分析HJ-1A/BCCD影像中不同地物的多种指数基础上,提出了改进暗目标法实现HJ-1A/BCCD影像的大气订正,该方法采用比值植被指数(RVI)、土壤调整植被指数(SAVI)和归一化水体指数(NDWI)的综合分析法确定暗像元自动提取,使之适用于环境减灾卫星CCD影像数据.为了客观地验证该方法的精度,本文选取地表平坦均一的敦煌校正场作为实验区,通过多次测量卫星过境时的地表反射率进行分析验证.
Atmospheric correction is an indispensable step in the quantitative research of remote sensing information. There are some mature methods at present. However, due to the band setting characteristics of HJ-1A / B satellite CCD camera, the conventional atmospheric correction method is not suitable for HJ-1A / BCCD images.On the basis of extensive analysis of many kinds of indices of different features in HJ-1A / BCCD images, this paper proposes atmospheric correction of HJ-1A / BCCD images by using improved dark target method. This method uses the ratio of vegetation index (RVI) , Soil adjusted vegetation index (SAVI) and normalized water body index (NDWI) to determine the automatic extraction of dark pixels, making it suitable for environmental disaster reduction satellite CCD image data.In order to objectively verify the accuracy of the method, this paper selects The Dunhuang calibration field, which is flat and uniform on the ground surface, is used as an experimental area, and the surface reflectance at the time of satellite transit is analyzed and verified.