第三章二遥感数据显示与基本分析.ppt

遥感图像可视化--最佳指数因子 图像数据的标准差越大所包含的信息量也越大，波段间的相关系数越小，表明各波段图像数据的独立性越高，信息的冗余度越小 * Here is an example of part of a Landsat “scene.” Note that all seven bands of data are in grey scale, even those that represent visible portions of the spectrum (blue, green or red reflected light), until we assign a color to the data. In this case we assigned the color blue to the band that represents blue light, the color green to the band representing green light, and red to the band of data representing red light. This produces what is called a “true color” image, akin to what one would see if they could ride atop one of these satellites (and used super-powered binoculars!) * Sometimes, to make things easier for our eyes to see, we use “false color.” One typical band combination for this is 4,3,2. Here band 4 (the near infra red) is assigned the color red. Band 3 (representing light reflecting in the red portion of the spectrum) is assigned the color green, and band 2 (what you and I would see as green) is assigned the color blue. Why do this? Recall that healthy vegetation reflects strongly in the near infra-red. If we color vegetation red, our eyes pick that out most clearly. In 4,3,2 combinations, the blue-green indicates the built environment (roads, pavement, buildings). Black or dark blue indicates water (or cloud shadows!) The choice of color to represent each wavelength range (band) doesnt matter as long as it aids interpretation, in other words, as long as it helps to understand the data represented in the image. 遥感数据可视化与基本分析 主要内容 遥感图像可视化 遥感图像基本分析量 遥感图像直方图及其应用 遥感图像可视化 遥感图像数据可视化 定义：以视觉方式探究数据和信息，从而获得对数据内涵的理解 方式： 二维空间：真彩色显示 三维空间：叠加数字高程模型 硬拷贝：打印机等输出 临时拷贝：显示影像 遥感图像显示的关键问题 创建亮度图--数据亮度值在计算机上的表现 保持亮度值和输出显示值的一一对应关系 遥感图像可视化--影像显示 灰度显示 RGB显示 颜色查找表 假彩色显示 5 6 7 4 3 2 1 3,2,1 Visible Infrared Green Data is shown as Green Blue Data is shown as Blue Red Data is shown as Red 5 6 7 4 3 2 1 4,3,2 Visible Infrared NIR Data is shown as Red Red Data is shown as Green Green Data is shown as B