论文部分内容阅读
WorldView-2卫星自2009年发射至今,已为用户提供了大量高性能的影像产品。与众多高分辨率卫星影像不同,WorldView-2有2个近红外波段,即近红外1(Near-infrared1,NIR1)和近红外2(Near-infrared2,NIR2),但目前这2个波段在应用上的区别并不清楚。因此,本文以福建省长汀县河田地区的马尾松林为例,采用NIR1和NIR2这2个近红外波段分别构建了3种植被指数(NDVI、ARVI和NDMVI),以探索二者在植被信息反演方面的差异。结果表明,NIR1构建的植被指数在马尾松林提取精度上高于NIR2,并具有更丰富的植被信息量。经统计可知,NIR1所构建的植被指数信息量比NIR2分别大8.0%(NDVI)、12.3%(ARVI)和7.3%(NDMVI);在反演植被覆盖度方面,NIR1也比NIR2具有更高的精度,其模拟的植被覆盖度与实际植被覆盖度的拟合度更高,误差更小。NIR1和NIR2所表现出的差异是因为马尾松在这2个近红外波段的光谱反射不同,其反射在NIR1的波长范围内达到最强,而在NIR2的波长范围内则出现了小幅下降。
Since its launching in 2009, WorldView-2 satellite has provided users with a large number of high-performance video products. Unlike many high-resolution satellite images, WorldView-2 has two near-infrared bands, Near-infrared1 (NIR1) and Near-infrared2 (NIR2), but these two bands are currently in use The difference is not clear. Therefore, taking Pinus massoniana forest in Hetian of Changting County of Fujian Province as an example, we constructed three vegetation indices (NDVI, ARVI and NDMVI) using NIR1 and NIR2, respectively, in order to explore the relationship between vegetation information Performance differences. The results showed that the vegetation index constructed by NIR1 was higher than that of NIR2 in Pinus massoniana plantation and had more abundant vegetation information. According to the statistics, the information of vegetation index constructed by NIR1 is 8.0% (NDVI), 12.3% (ARVI) and 7.3% (NDMVI) higher than that of NIR2 respectively. NIR1 is also higher than NIR2 in inverting vegetation coverage Accuracy, the simulated vegetation coverage and the actual vegetation coverage of the higher degree of fit, the error is smaller. The difference between NIR1 and NIR2 is due to the different spectral reflections of the masson pine in the two NIR bands, the strongest reflection in the NIR1 wavelength range and the slight decrease in the NIR2 wavelength range.