论文部分内容阅读
植物冠层色素含量与氮素含量具有高度的相关性,是农业遥感中的关键研究因素。本研究的主要目的是:(1)对比偏最小二乘回归和支持向量机两种建模方法对枣树冠层色素的预测精度;(2)构建基于高光谱数据的枣树冠层色素含量定量反演模型,为枣树冠层色素含量的快速、无损、廉价、环保的测定提供一定的理论依据和技术支持。相关性分析结果表明,枣树冠层色素与高光谱数据之间具有较好的相关性,但叶绿素、叶绿素a要优于叶绿素b和类胡萝卜素。独立样本对模型的预测性能检验结果表明,偏最小二乘回归和支持向量机均能有效的估算枣树色素含量,但不同色素的偏最小二乘回归模型和支持向量机模型的预测精度存在一定的差异,叶绿素和类胡萝卜素的支持向量机模型的预测精度要高于偏最小二乘回归模型,而叶绿素a和叶绿素b则相反。比较不同色素的最佳反演模型的预测精度表明,叶绿素、叶绿素a和类胡萝卜素的预测精度要优于叶绿素b,前三者的决定系数大于0.8,残余预测误差高于2.0,平均相对误差低于13%,而叶绿素b的对应值分别为0.60%,20.79%和1.79%。
Plant canopies pigment content and nitrogen content has a high correlation, is the key factor in agricultural remote sensing. The main objectives of this study are: (1) to compare the prediction accuracy of jujube canopy pigment with two methods of partial least-squares regression and support vector machine; (2) to construct quantitative analysis of jujube canopy pigment content based on hyperspectral data Play a model for the jujube canopy pigment content fast, non-destructive, cheap, environmentally friendly determination to provide a theoretical basis and technical support. Correlation analysis showed that there was a good correlation between canopy pigment and hyperspectral data, but chlorophyll a and chlorophyll a were superior to chlorophyll b and carotenoid. The test results of the independent samples on the model show that both PLSVM and SVM can effectively estimate the pigment content in jujube tree, but there is a certain degree of prediction precision in partial least squares regression model and support vector machine model The prediction accuracy of SVM models of chlorophyll and carotenoid is higher than that of PLS model, while the chlorophyll a and chlorophyll b are the opposite. The prediction accuracy of the best inversion model comparing different pigments shows that the prediction accuracy of chlorophyll, chlorophyll a and carotenoid is better than that of chlorophyll b, the former three are greater than 0.8, the residual prediction error is higher than 2.0, the average relative error Below 13%, while the corresponding values for chlorophyll b were 0.60%, 20.79% and 1.79%, respectively.