【作者】 郑伟；

【导师】 曾志远；

【作者基本信息】 南京师范大学 ， 地图学与地理信息系统， 2005， 硕士

【摘要】 本论文是导师曾志远先生主持的国家自然科学基金项目“SPOT图像信息提取与土地利用/土地覆盖识别与监测研究”(批准号:40371053)资助下,为使不同时间或不同年份的遥感图像具有数值上的可比性,实现利用遥感图像对土地利用/土地覆盖内容的更精细识别和对它们的定量化动态监测,对遥感图像大气校正方法进行了研究。 目前,在遥感图像大气校正方法中,有一些精度较高的大气校正模型,但是这些模型需要较多的运行参数,且其中有些参数很难获得,而如果用模型中默认的参数,又使得模型的运行精度下降。 本论文在对TURNER大气校正模型分析的基础上,对TURNER大气校正模型中的一些参数进行修正,特别是对关键参数气溶胶光学厚度的计算模型进行修正,使其较适合中国地区,得到修正的TURNER大气校正模型。利用修正的TURNER大气校正模型对南京地区的SPOT5和ETM+图像进行大气校正,并与已有的TURNER模型、6S大气校正模型对图像的校正结果以及利用ASD光谱仪实测的样点光谱数据进行比较分析,最终证明修正的TURNER模型大气校正效果好,获得运行相对简单、需要较少的大气校正参数、满足一定精度的大气校正模型。同时,为了使给模型能更方便的应用,把该模型用C++语言实现,得到遥感图像大气校正实用程序。最后,利用修正的TURNER大气校正模型对另外三幅南京地区的遥感图像大气校正,共获得五幅遥感图像大气影响波段差异系数,并与山地阴影主体大气校正方法得出的大气影响波段差异系数进行比较分析,结果很相近,说明山地阴影主体大气校正方法是比较科学的。更多还原

【Abstract】 The research is funded by the national natural science fund project titled Information Extraction of SPOT Satellite Remote Sensing Image and Dynamic Monitoring of Land Use and Land Cover(sanction number: 40371053 ).In order to ensure that images of different time are comparable and realize Dynamic Monitoring of Land Use and Land Cover based on satellite remote sensing image, methods of atmospheric correction of satellite remote sensing image are studied.At present, there are some methods of atmospheric correction with acceptable accuracy .But, the operation of these methods need a lot of parameters which are usually difficult to get. If default parameters of these methods are applied, it will result in unacceptable accuracy.In this paper, TURNER model of atmospheric correction is modified , in particularly, the model of the aerosol optical depth of the key parameter is modified to be more suitable for the aerosols in China. The SPOT5 and ETM+ satellite remote sensing images of the area of NanJing city in China are corrected by the modified model respectively. To analyze the accuracy of modified model , TURNER model and 6S atmospheric correction model are applied to correct these two scenes satellite remote sensing images and Pseudo-simultaneous ground’s samples reflectance measurements are available. After comparing with TURNER model ,6S model and reflectance measurements and analyzing the accuracy , the applicable and efficient atmospheric correction method with few and facile parameters is got. At the same time, the modified TURNER model is realized by C++ language and the applied programme is available. At last, the orther three scenes satellite remote sensing images are corrected by the modified TURNER model and get the atmospheric influence band coefficients.The atmospheric influence band coefficients got by Main Mountain Shadow Histogram method are compared with the ones got by modified TURNER model. The results are similar. This prove that Main Mountain Shadow Histogram method is reasonable-更多还原