目录
摘要Ⅱ
关键词Ⅱ
AbstractⅡ
引言
引言1
1材料与方法1
技术路线 1
1.2 实验设计 2
1.3 数据的获取及测定 2
1.3.1 主动观测传感器获取光谱数据2
1.3.2 农学参数的获取2
1.3.3 室内化学分析数据的获取3
1.4 研究方法3
1.4.1 光谱指数及时间驱动因子计算3
1.4.2 光谱植被指数重构方法4
1.4.3 植被指数与氮营养指数的关系4
2 结果与分析5
2.1 不同氮水平对植被指数动态的影响5
2.2 不同时序植被指数拟合方法比较分析5
2.2.1 基于AGDD的时序植被指数曲线的拟合方法研究5
2.2.2 基于时序植被指数曲线的拟合模型7
2.3 植被指数NDRE和NDVI与农学参数的定量关系 8
2.4 基于时序植被指数的农学参数的定量关系 9
3 讨论和分析10
致谢10
参考文献11
时序植被指数不同构建方法的比较与评估
摘要
研究冬小麦冠层时序植被指数的动态变化规律并基于其监测植株长势，为田间实时、准确获取作物长势信息提供有效的技术手段[13]。本研究于20172018年在江苏省兴化市万亩粮食产业园开展不同氮肥水平的田间小区试验，利用主动冠层传感器RapidSCAN CS45获取归一化植被指数（Normalized difference vegetation index, NDVI）和归一化红边植被指数（Normalized difference red edge, NDRE），并在关键生育期取样测定植株叶面积指数（Leaf area index, LAI）、叶片生物量（Leaf dry matter, LDM）、叶氮积累量（Leaf nitrogen accumulation, LNA）等指标。基于时间驱动因子累积生长度日数（Accumulated Growing Degree Days, A *51今日免费论文网|www.jxszl.com +Q: ￥351916072$ 
GDD）的时序拟合方法（双Logistic函数、傅里叶变换、三次函数、高斯函数）重构时序植被指数，并基于其时序曲线提取特征参数，分析其与农学参数的相关关系。结果表明,基于时序植被指数NDRE和NDVI的四种重构方法中，双Logistic函数拟合效果较好，其R2最高，RMSE最低；其次是三次函数和傅里叶变换。NDRE和NDVI与叶面积指数和叶氮积累量的关系较好，R2大于0.7，其与叶片生物量的关系较差，R2小于0.6。基于时序NDRE和NDVI曲线的最大值和NDREsum、NDVIsum，在灌浆期与LAI相关关系较好，在孕穗期与LDM相关性较好，在拔节期与LNA相关性较好。基于时序NDRE曲线的ΔNDRE与LAI、LDM、LNA的相关性较时序NDVI曲线的ΔNDVI好。基于冠层时序植被指数的ΔNDRE和ΔNDVI都具有良好的监测小麦植株长势的潜力，研究结果为田间进行实时、准确的监测作物长势提供了技术支持。
Comparison and Evaluation of Different Construction Methods of Time Series Vegetation Index
ABSTRACT
To study the dynamic change rule of winter wheat canopy temporal sequence vegetation index and monitor plant growth based on it, so as to provide effective technical means for obtaining crop growth information in real time and accurately in the field.In this study, field community experiments with different nitrogen fertilizer levels were conducted in xinghua grain industrial park of 10,000 mu in jiangsu province from 2017 to 2018. Active canopy sensor RapidSCAN cs45 was used to obtain Normalized difference vegetation index (NDVI) and Normalized difference red edge (NDRE).Leaf area index (LAI), Leaf dry matter (LDM), Leaf nitrogen accumulation (LNA) and other indicators were measured in the critical growth period. Based on the timedriven AGDD timing sequence fitting method (double Logistic function, Fourier transform, cubic function, gaussian function), the time sequence vegetation index was reconstructed, the curve characteristic parameters were extracted, and the correlation between them and agricultural parameters was analyzed. The results showed that among the four reconstruction methods based on NDRE and NDVI, the fitting effect of the bilogistic function was better, with the highest R2 and the lowest RMSE.Then you have the cubic function and the Fourier transform.Based on the maximum values of NDRE and NDVI curves, NDREsum and NDVIsum at the grouting stage had a good correlation with LAI, LDM at the starting stage and LNA at the jointing stage.Based on timeseries NDRE curve Δ NDRE and LAI, the correlation of LDM, LNA a timeseries NDVI curve Δ NDVI is good.Based on canopy temporal vegetation index Δ NDRE and Δ NDVI has a good monitoring the potential of wheat plant growing, realtime and accurate results for the field monitoring crop condition provides the technical support.

原文链接：http://www.jxszl.com/nongxue/zwbh/606176.html