植物叶片叶绿素含量及分布是植物营养信息表达的一个重要指标.以茶树为研究对象,利用高光谱技术分析茶树叶片中叶绿素含量及其分布.通过采集茶树鲜叶的高光谱图像,利用7种不同的算法从高光谱数据中提取相应的特征参数,并根据特征参数和叶绿素含量的参考测量值分别拟合出相应的预测模型.结果显示,二次土壤调节植被指数算法提取的特征参数最佳,预测模型校正集和预测集的相关系数R分别为0.843 3和0.832 3,最小均方根误差分别为9.918和8.601.最后根据预测模型估计叶片上任意像素下叶绿素的含量,并通过伪彩手段描述叶片中叶绿素含量的分布.研究结果表明,利用高光谱成像技术分析茶树叶片中叶绿素含量及其分布是可行的.%Chlorophyll content and distribution in plant's leaves is an important index in estimation of plant nutrition information. In the present work, chlorophyll content and distribution in tea plant's leaves were measured by hyperspectral imaging technique. First, hyperspectral image data were captured from tea plant's leaves; then seven kinds of algorithms were used to extract the characteristic parameters from hyperspeetral image; finally, seven fitted models were developed using the characteristics vectors and the reference measurements of chlorophyll contents respectively. Experimental results showed that the MSAVI2 model is superior to other models, and the results of the MSAVI2 model was achieved as follows: R=0. 843 3 and RMSE=9. 918 in the calibration set; R＝0. 832 3 and RMSE＝8. 601 in the prediction set. Finally, the chlorophyll content of each pixel in image was estimated by the fitted model, and the distribution of chlorophyll content in the tea plant's leaf was described by pseudo-color map. This study sufficiently demonstrated that the chlorophyll content and distribution in tea leaf can be measured by hyperspectral imaging technique.
展开▼
