棉花全生育期叶片SPAD值的遥感估算模型

发布时间：2018-10-18 15:13

【摘要】：叶绿素含量是评估棉花生长状况的重要参数,估算叶绿素含量对于棉花生长监测具有重要意义。以渭北旱塬区种植的棉花为试验材料,测量全生育期棉花叶片SPAD值与冠层反射率光谱,将原始高光谱反射率、一阶微分光谱反射率、不同波段组合的遥感光谱参数分别与SPAD值做相关性分析,用传统回归分析方法构建五种重要光谱参数的SPAD值预测模型,同时,采用PLSR方法建立全生育期SPAD值的估算模型。最后对模型进行检验,筛选出精度最高的模型。建模结果表明,基于多种光谱参数的全生育期PLSR预测模型精度最高、预测效果最好,估算模型的决定系数R~2为0.733,验证模型R~2为0.737。PLSR方法建立的多光谱参数的SPAD值估算模型预测效果显著,利用高光谱技术对棉花SPAD值进行监测,可为全生育期棉花长势遥感监测提供依据。
[Abstract]:Chlorophyll content is an important parameter to evaluate cotton growth, and estimation of chlorophyll content is of great significance for cotton growth monitoring. Using cotton planted in Weibei dryland as experimental material, the SPAD value and canopy reflectance spectrum of cotton leaves during the whole growing period were measured. The original hyperspectral reflectance and the first order differential spectral reflectance were measured. The correlation between the spectral parameters of remote sensing and the SPAD value of different band combinations was analyzed, and the SPAD prediction model of five important spectral parameters was constructed by using the traditional regression analysis method. At the same time, the PLSR method was used to establish the estimation model of the SPAD value in the whole growth period. Finally, the model is tested and the model with the highest precision is screened out. The modeling results show that the PLSR prediction model based on multiple spectral parameters has the highest precision and the best prediction effect. The determination coefficient of the estimation model is 0.733, and the prediction effect of the SPAD model for estimating the multispectral parameters established by the 0.737.PLSR method is remarkable. The hyperspectral technique is used to monitor the SPAD value of cotton. It can provide the basis for cotton growth remote sensing monitoring during the whole growth period.
【作者单位】：西北农林科技大学资源环境学院;
【基金】：国家高技术研究发展计划(863计划)项目(2013AA102401-2)
【分类号】：S562;TP79

【相似文献】