基于地形辐射校正的植被覆盖参数遥感反演

发布时间：2018-06-30 02:07

本文选题：遥感影像 + 地形辐射校正　；参考：《山东农业大学》2017年硕士论文

【摘要】：遥感作为一种远距离非接触式获取地表信息的方法,越来越受到人们的重视。在遥感影像的地物光谱信息分析中,定量遥感一直是人们研究的一个重要方向,由于卫星是在高空获取地面图像,不可避免的受到电磁波、大气、地形等因素的干扰,其中,由于地形起伏造成的地物光谱信息的改变是影响影像质量的重要原因之一。地形起伏使太阳光易被山体遮挡,导致背向阳光的坡面不能反射太阳的直射光线,地物纹理特征被削弱,影响了影像的判读,降低了地物分类精度等诸多遥感分析的过程。因此,基于地形辐射校正的影像光谱信息恢复是以遥感手段准确获取地表真实光谱信息不可或缺的一步。本文首先介绍地形辐射校正的研究背景和国内外研究现状,并用ENVI软件对影像进行大气校正、几何校正等预处理。然后利用MATLAB软件对研究所用的地形辐射校正方法(C校正、SCS+C校正、Minnaert-SCS校正、Teillet-回归校正、VECA校正)进行代码编写和处理。在Minnaert-SCS校正模型的基础上通过引入半经验参数C对其进行改进,避免在太阳入射角余弦值cosi过小时的过度校正现象,并编码处理比较其校正效果。最后提取出RVI、NDVI、EVI、MSAVI四种植被指数并作比较分析,利用像元二分法分别计算这四种植被指数下的植被覆盖度,分析地形辐射校正对植被覆盖度反演的影响。通过实验处理和比较发现,改进的校正方法可以提高地形效应下阴影区域和低亮度区域的校正效果,能消除大部分反射率值的差异,通过目视比较和统计数据分析发现,改进的模型取得了较好的校正效果;另一方面地形辐射校正对基于EVI计算的植被覆盖度起到了改善作用,对基于RVI、NDVI和MSAVI计算的植被覆盖度提升较小,建议在使用EVI估算多山区域的植被覆盖度时进行地形辐射校正。
[Abstract]:As a long distance non-contact method to obtain surface information, remote sensing has attracted more and more attention. Quantitative remote sensing has always been an important research direction in the spectral information analysis of ground objects in remote sensing images. Because the satellite acquires ground images at high altitude, it is inevitably disturbed by electromagnetic waves, atmosphere, topography and other factors, among which, The change of spectral information caused by topographic fluctuation is one of the important factors that affect the image quality. The relief of the terrain makes the sunlight easily obscured by the mountain body, which can not reflect the direct ray of the sun on the slope of the back sun, and the texture feature of the ground object is weakened, which affects the interpretation of the image and reduces the process of remote sensing analysis such as the precision of the classification of the ground objects and so on. Therefore, the restoration of image spectral information based on terrain radiation correction is an indispensable step to accurately obtain the true spectral information of the surface by remote sensing. This paper first introduces the background of topographic radiation correction and the present research situation at home and abroad, and uses ENVI software to preprocess the image such as atmospheric correction, geometric correction and so on. Then, the method of terrain radiation correction (C correction SCS C correction and Minnaert-SCS correction / Teillet-regression correction Veca correction) used in the research is compiled and processed by MATLAB software. Based on the Minnaert-SCS correction model, the semi-empirical parameter C is introduced to improve the correction, to avoid the over-correction of the cosine value cosi at the solar incidence angle, and to compare the correction effect with the coding process. Finally, four vegetation indices were extracted and compared with each other. The vegetation coverage under the four vegetation indices was calculated by pixel dichotomy, and the effect of topographic radiation correction on vegetation coverage inversion was analyzed. Through experimental processing and comparison, it is found that the improved correction method can improve the correction effect of shadow area and low luminance area under terrain effect, and eliminate the difference of most reflectivity values. Through visual comparison and statistical data analysis, it is found that, On the other hand, topographic radiation correction can improve vegetation coverage based on EVI, and improve vegetation coverage based on RVI NDVI and MSAVI. It is suggested to use EVI to estimate vegetation coverage in mountainous areas.
【学位授予单位】：山东农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P237;Q948

【参考文献】