不同光谱植被指数反演冬小麦叶氮含量的敏感性研究

发布时间：2019-06-24 14:36

【摘要】：【目的】氮素是作物生长发育过程中最重要的营养元素之一,研究叶氮含量反演的有效光谱指标设置,为应用高光谱植被指数反演作物叶氮含量,以及作物的实时监测与精确诊断提供重要依据。【方法】以冬小麦为例,选取涵盖冬小麦全生育期不同覆盖程度225组冠层光谱与叶氮含量数据,通过遥感方法建立模型,模拟了不同光谱指标,即中心波长、信噪比和波段宽度对定量模型的影响,通过模型精度评价指标决定系数(coefficient of determination,R~2)、根均方差(root mean square error,RMSE)、平均绝对误差(mean absolute error,MAE)、平均相对误差(mean relative error,MRE)和显著性检验水平(P0.01)确定最优模型及最佳指标,分析光谱指标对叶氮含量定量模型反演的敏感性和有效性。【结果】反演冬小麦叶氮含量的最佳植被指数为MTCI_B,与实测叶氮含量的相关性最好(R~2=0.7674,RMSE=0.5511%,MAE=0.4625%,MRE=11.11个百分点,且P0.01),对应的最佳指标为中心波长420 nm、508 nm和405 nm,波段宽度1 nm,信噪比大于70 DB;高覆盖状况反演的最优指数为RVIinf_r(R~2=0.6739,RMSE=0.2964%,MAE=0.2851%,MRE=6.44个百分点,且P0.01),最优中心波长为826 nm和760 nm;低覆盖状况反演的最优指数为MTCI(R~2=0.8252,RMSE=0.4032%,MAE=0.4408%,MRE=12.22个百分点,且P0.01),最优中心波长为750 nm、693 nm和680 nm;应用最适于高低覆盖的植被指数RVIinf_r和MTCI构建的联合反演模型(R~2=0.9286,RMSE=0.3416%,MAE=0.2988%,MRE=7.16个百分点,且P0.01),明显优于最佳单一指数MTCI_B;模拟Hyperion和HJ1A-HSI传感器数据,联合反演模型精度(R~2为0.92—0.93,RMSE在0.37%—0.39%,MAE为0.285%左右,MRE约为7.00个百分点)明显优于单一植被指数反演精度(R~2为0.79—0.81,RMSE为0.63%—0.66%,MAE为0.455%左右,MRE约为10.90个百分点)。【结论】利用高光谱植被指数可有效实现作物叶氮含量反演,作物叶氮含量定量反演对不同光谱指标—中心波长、信噪比和波段宽度,具有较强敏感性。应用多指数联合反演模型,可显著提高反演精度,并且联合反演模型在不同高光谱传感器下有一定普适性。
[Abstract]:[Objective] N is one of the most important nutrient elements in the process of crop growth and development, and the effective spectral index of the inversion of leaf nitrogen content is set up to provide an important basis for the application of high-spectral vegetation index to the inversion of crop leaf nitrogen content, as well as real-time monitoring and accurate diagnosis of crops. [Method] Taking winter wheat as an example, the data of the canopy spectrum and leaf nitrogen content of 225 groups covering the whole growth period of the winter wheat were selected, and the model was established by the remote sensing method. The effects of the central wavelength, the signal-to-noise ratio and the band width on the quantitative model were simulated. The optimal model and the best index were determined by the model accuracy evaluation index determination coefficient (R ~ 2), the root mean square error (RMSE), the mean absolute error (MAE), the mean relative error (MRE) and the significance test level (P0.01), and the sensitivity and the effectiveness of the spectral index to the inversion of the leaf nitrogen content quantitative model were analyzed. [Results] The best vegetation index for the inversion of the leaf nitrogen content of winter wheat is MTCI _ B, and the correlation with the measured leaf nitrogen content is the best (R ~ 2 = 0.7674, RMSE = 0.5511%, MAE = 0.4625%, MRE = 11.11%, and P 0.01), the corresponding best index is the center wavelength of 420 nm,508 nm and 405 nm, the band width is 1 nm, the signal-to-noise ratio is greater than 70 DB, the optimal index for the inversion of the high coverage condition is RVIinf r (R ~ 2 = 0.6739, RMSE = 0.2964%, MAE = 0.2851%, MRE = 6.44%, and P0.01), the optimal center wavelength is 826 nm and 760 nm; the optimal index for low coverage condition inversion is MTCI (R ~ 2 = 0.8252, RMSE = 0.4032%, MAE = 0.4408%, MRE = 12.22%, and P0.01) The RMSE is 0.37% (0.39%, MAE is about 0.285%, and the MRE is about 7.00%), which is obviously superior to that of single vegetation index (R ~ 2 is 0.79, 0.81, RMSE is 0.63%, 0.66%, MAE is about 0.455%, and MRE is about 10.90%). [Conclusion] Using the high-spectral vegetation index, the inversion of the crop leaf nitrogen content and the quantitative inversion of the leaf nitrogen content of the crop have a strong sensitivity to the central wavelength, the signal-to-noise ratio and the band width of different spectral targets. The multi-index joint inversion model can be used to significantly improve the inversion precision, and the joint inversion model has a certain universality under different high-spectral sensors.
【作者单位】：中国科学院遥感与数字地球研究所;中国地质大学地球科学与资源学院;
【基金】：国家自然科学基金(41371359,41671362) 广西空间信息与测绘重点实验室开放基金(151400727) 高分辨率对地观测系统重大专项(30-Y30B13-9003-14/16,11-Y20A40-9002-15/17)
【分类号】：S512.11;S127

【相似文献】