| 李梦竹,刘国顺,贾方方,杨艳东.旺长期烤烟对不同程度干旱胁迫的光谱响应[J].干旱地区农业研究,2017,35(3):164~171 |
| 旺长期烤烟对不同程度干旱胁迫的光谱响应 |
| Spectral response characteristics of flue-cured tobacco under different drought stress at the vigorous growth stage |
| |
| DOI：10.7606/j.issn.1000-7601.2017.03.26 |
| 中文关键词: 烤烟叶片 旺长期 水分含量 光谱指数 估算模型 |
| 英文关键词:flue-cured tobacco leaves the vigorous growth stage water content spectral index estimating model |
| 基金项目:国家烟草专卖局特色优质烟叶重大专项浓香型特色优质烟叶开发项目(110201101001(TS-01))；河南中烟工业有限责任公司科技项目(ZW2014005) |
|
| 摘要点击次数: 1130 |
| 全文下载次数: 546 |
| 中文摘要: |
| 基于不同时长和程度的干旱胁迫试验，采用ASD光谱仪，研究了不同干旱胁迫处理下不同基因型烤烟叶片水分含量与光谱特征的变化规律，分析不同烤烟叶片水分指标FMC（叶片相对含水量）、LEWT（叶片等效水厚度）与光谱特征参数间的相关关系，构建烤烟叶片水分指标的高光谱特征参数模型。结果表明：不同程度干旱胁迫下烤烟叶片含水量和光谱反射率均随干旱程度的加重而降低，在不同干旱时长下，FMC的变化并不明显，而LEWT对其较为敏感，说明不同干旱胁迫处理间EWT的差异要比FMC显著，不同基因型烤烟品种表现一致，表明LEWT比FMC更适合反映旺长期烤烟叶片水分状况。利用光谱参数建立FMC和LEWT的一元及多元线性模型和BP神经网络模型中，均以BP神经模型网络模型效果最好，其模型R2（决定系数）分别为0.8650、0.9464，RMSE（均方根误差）分别达到0.0049、0.0047，表明模型的精度和稳定性均较好。 |
| 英文摘要: |
| Drought directly affects the production and quality of flue-cured tobacco in particular the water critical period. The objective of this paper was to investigate the water contents of flue-cured tobacco leaves and hyperspectral features under different drought duration and drought stress by ASD spectrometer, with a model of the flue-cured tobacco leaf water indexes. The results indicated that the leaf water content and spectral reflectance under different treatments were reduced with the increase of drought degree. Under different drought duration, the leaf equivalent water thickness (LEWT) was sensitive, while fuel moisture content (FMC) was stable, suggesting that LEWT may be a suitable indicator of the water content of flue-cured tobacco leaves. The monadic linear model, multivariate linear model and BP neural network model were assembled for FMC and LEWT based on spectral parameters, the BP neural network model showed the best optimal effect, with the R2 being 0.8650, 0.9464 and RMSE being 0.0049, 0.0047, respectively. This model had high precision and excellent stability, which may provide guidance for monitoring the drought disaster and the development of digital tobacco using hyperspectral remote sensing technology during the vigorous growth stage. |
| 查看全文 查看/发表评论 下载PDF阅读器 |