遥感图像信息容量与植被指数的耦合关系研究

Coupling relationship between remote sensing image information capacity and a variety of vegetation index

遥感信息容量能表征地表复杂度,为深入探究信息容量与植被之间关系,研究对植被盖度、NDVI、RVI等5种植被指数的均值、均方差与信息容量做了多种回归分析。结果表明:植被覆盖度均值、植被指数均方差与信息容量有明显相关性,其中植被盖度均方差与信息容量相关性最大,确定系数R^2为0.851 8;相关系数最大R^2为0.8642,说明遥感信息容量与植被指数、植被盖度之间有明显的相关性,证明了信息容量能有效地表征地表覆盖的变化。

The remote sensing image information capacity is a kind of evaluation model which be able to character- ize the surface complexity, and its calculated value size is closely related to the number of image gray levels. Be- sides, remote sensing image information capacity is also sensitive to the change of image grey value. Vegetation is the main factor of surface cover, and different vegetation types have different gray values, thereby affecting the value of the DN of the image, so the fractional vegetation cover is the main factor of affecting the vegetation cover gray complexity. Information capacity is calculated based on image DN value image quality evaluation criteria that can accurately and effectively express gray structure information. Therefore changes in vegetation will affect the image information capacity, and vegetation index is an important factor in the study of vegetation in remote sensing. The pa- per selected 19 samples in the same period of the image to calculate the information capacity of these samples and a variety of vegetation index, in order to further research the coupling between information capacity and fractional veg- etation cover. The paper mainly studies the fractional vegetation cover, NDVI, RVI and other two vegetation index- es. At the same time, it also makes various correlation analyses among fractional vegetation cover, the mean and standard deviation of the five vegetation indexes and information capacity. The result of research shows that informa- tion capacity has less correlation with the mean of fractional vegetation cover and vegetation index and information capacity, but quite related to standard deviation. Among that, standard deviation serves the largest correlation with information capacity as that： R2 =0.824 5. At the same time, the correlation analysis is made between standard devi- ation of fractional vegetation cover and information capacity, the maximum correlation coefficient between them is R2 =0.864 2. Finally, the study not only shows that the remote sensing image information capacity has obvious corre- lation with these several types of vegetation indexes and fractional vegetation cover, but also can effectively charac- terize the changes of surface structure.