论文部分内容阅读
在对毛竹林叶片高光谱反射率数据进行小波变换的基础上,寻找和确定最佳的小波植被指数反演毛竹林叶片的净光合速率(P_n).结果表明:理想的高频小波植被指数反演得到的P_n精度高于低频小波植被指数和光谱植被指数,其中,由小波分解第一层高频系数构建的归一化植被指数、比值植被指数和差值植被指数与P_n之间的相关性最好,R~2为0.7,均方根误差(RMSE)较低,为0.33;而低频小波植被指数反演P_n的精度低于光谱植被指数.由各层理想小波植被指数所构建的多元线性模型反演得到毛竹叶片P_n与实测P_n之间具有显著的相关关系,R~2为0.77,RMSE为0.29,且精度明显高于基于光谱植被指数所构建的多元线性模型.与光谱植被指数反演毛竹P_n的敏感波段仅局限于可见光波段相比,小波植被指数探测的敏感波长范围更广,包含了可见光及多个红外波段.高光谱数据在经过小波变换后能够发现更多反映毛竹P_n的细节信息,且整体反演精度比原始光谱有了显著提高,研究结果为基于高光谱遥感反演植被P_n提供了一种新的可选方法.
Based on wavelet transform of Hyperspectral reflectance data of Phyllostachys heterocycla cv. Pubescens, the optimal PWI was used to retrieve the net photosynthetic rate (P_n) of Phyllostachys pubescens leaves.The results showed that the ideal high frequency wavelet vegetation index The accuracy of P_n obtained by the above method is higher than that of low frequency wavelet and vegetation index, in which the correlation between normalized vegetation index, ratio vegetation index and difference vegetation index and P_n constructed by the first layer high frequency coefficients of wavelet decomposition The best RMSE was 0.33 for R ~ 2 and 0.33 for root mean square error (RMSE). The accuracy of inversion of P_n by low-frequency wavelet vegetation index was lower than that of spectral vegetation index.The multivariate linearity The results showed that there was a significant correlation between P_n and measured P_n, which was 0.77 for R ~ 2 and 0.29 for RMSE, and the precision was significantly higher than that of multivariate linear model constructed based on spectral vegetation index.Compared with spectral vegetation index Compared with the visible wavelength band, the sensitive wavelength range of P_n is wider, including the visible light and multiple infrared bands. The hyperspectral data after wavelet transform Possible to find more detailed information reflecting P_n bamboo, and overall accuracy has been significantly improved inversion than the original spectrum, the inversion results for the vegetation P_n provides a new alternative method from Hyperspectral.