【作者】 王随霞；

【导师】 张行南； 朱辰华；

【作者基本信息】 河海大学 ， 水文学及水资源， 2006， 硕士

【摘要】 本文是在黄河防汛科技项目“MODIS数据在黄河防洪及防凌中的应用研究”的支持下展开的,是为研究MODIS数据的黄河洪水监测方法而进行的理论和试验研究。 洪水灾害是当今世界上造成损失最大的自然灾害。在洪灾发生时,快速、准确地提取水体信息,确定洪水淹没范围,为防洪决策提供科学依据,具有十分重要的意义。MODIS遥感数据以其高空间分辨率、高时间分辨率和高光谱等显著优点,在洪水动态监测中具有广阔的前景。 本文在详细分析了近年来国内外遥感洪水监测研究进展的基础上,以黄河下游花园口至孙口河段为研究区,根据黄河水体在MODIS数据上的光谱特征及空间特征,建立了基于MODIS数据的黄河下游水体信息提取模型,并讨论了含沙量等因素对模型参数的影响。论文研究的内容主要涵盖以下几个方面: (1) 以突出黄河水体信息为目的,对MODIS L1B数据进行了简单处理,主要包括:波段选择、“蝴蝶结”去除、几何校正、辐射校正及数据融合等。 (2)分析了黄河水体在MODIS影像上的光谱表现特征及与其它背景地物在光谱特征上的差异,分别采用单波段阈值法和多光谱混合分析法对黄河水体进行了提取实验研究,并对比分析了两种方法的优缺点,指出多光谱混合分析法更适用于MODIS数据的黄河水体提取。 (3)建立了基于光谱特征和空间特征的黄河水体信息提取模型,并通过洪水监测实例分析了含沙量等因素对模型参数的影响。更多还原

【Abstract】 Supported by flood prevention technology project of the Yellow River— The application research of the MODIS data in the Yellow River flood control and anti-ice, this article presents the theoretic and experimental research on flood monitoring methods of the Yellow River using MODIS data.Flood disaster is the world’s largest natural disaster losses. When floods occur, it is of great significance to quickly and accurately extract water body information, identify the scope of the floods submerged, and provide scientific basis for flood control decision-making. With high spatial resolution, high temporal resolution, high spectral and other significant advantages, MODIS data has broad prospects in dynamic flood monitoring.Based on the analysis of research progress of flood monitoring using remote sensing in recent years, this paper deals with the methods of monitoring flood with the river section Huayuankou to Sunkou of the Yellow River as study area. According to the spectral characteristic and the spatial characteristic of the Yellow River water body in MODIS image, we established the water body information extraction model of the Yellow River based on the MODIS data, and discussed the influence of sediment concentration and other factors to the model parameters. The content covers the following sections:(1) Taking highlights the Yellow River water body information as the goal, simple treatments are carried on to the MODIS L1B data, mainly includes: the band choice, the "bow-tie" removes, the geometric rectification, the radiation adjustment and the data fusion and so on.(2) The spectral characteristic difference between the Yellow River water body and other background things in MODIS image has analyzed. The single band threshold algorithm and the multi-spectral integrated analysis method were separately carried on to withdrawn the Yellow River water body as experimental study, and were contrasted to analyze two methods’ good and bad points. The multi- spectral integrated analysis method was pointed out suitable to the water body extraction ofYellow River with MODIS data.(3) The water body information extraction model of Yellow River based on the spectrum characteristic and the spatial characteristic was established. According to some flood monitor examples, the influence of the sediment concentration and other factors to the model parameters was analyzed.更多还原