基于SVR算法的小麦冠层叶绿素含量高光谱反演

Chlorophyll content inversion with hyperspectral technology for wheat canopy based on support vector regression algorithm

为给小麦的长势监测与农艺决策提供科学依据,利用高光谱技术实现了小麦冠层叶绿素含量的估测。通过分析18种高光谱指数对叶绿素的估测能力,筛选出可敏感表征叶绿素含量的指数REP,利用地面光谱数据为样本集,以最小二乘支持向量回归(least squares support vector regression,LS-SVR)算法建立了小麦冠层叶绿素含量反演模型,其校正决定系数C-R2与预测决定系数P-R2分别为0.751与0.722,在各指数中反演精度最高。进一步分析表明,REP对叶绿素含量以及LAI值较高与较低的样本均具备良好的预测能力,可有效避免样本取值范围以及冠层郁闭度等因素对叶绿素含量估测的影响。利用LS-SVR反演模型完成了OMIS影像叶绿素含量的遥感填图,并以地面实测值进行检验,其拟合模型R2与RMSE值分别为0.676与1.715。结果表明,高光谱指数REP所建立的LS-SVR模型实现了叶绿素含量的准确估测,可用于小麦叶绿素含量信息的快速、无损获取。

In order to provide scientific basis for wheat growth monitoring and agronomic decision-making, the wheat canopy chlorophyll content was estimated by using hyperspectral technology in this paper. Eighteen kinds of hyperspectral indices were comparative analyzed. The index REP, which could respond wheat canopy chlorophyll content sensitively, was selected. The inversion model of wheat canopy chlorophyll content was then built by using the field spectra as the training samples and the least squares support vector regression （LS-SVR） algorithm as the modeling method, with the calibration R-square and prediction R-square 0.751 and 0.722, respectively, indicating the accuracy of estimation predicted by REP was highest in all indices. Further more, the prediction accuracy of REP was least sensitive to the change of chlorophyll content and LAI values among 18 indices and therefore least affected by the range of sample values and canopy density when used to estimate the chlorophyll content of wheat canopy. Using the inversion model, the remote sensing mapping for OMIS image was accomplished. The inversion and measured values were then compared by the method of regression fitting. The R-square and RMSE of the fitting model was 0.676 and 1.715, respectively, indicating the similarity between the inversion value and measured value was high. The result showed that it was feasible to estimate chlorophyll content accurately by using hyperspectral index PEP to build a LS-SVR inversion mode. Therefore, this method proposed can be used as a rapid and non-destructive method for getting wheat chlorophyll content.