论文部分内容阅读
快速准确的粮食作物产量估算对于国家制订粮食政策和农业可持续发展具有重要意义。利用地面高光谱遥感的优点,获取作物冠层的精细光谱,并根据植被绿峰、红边、水汽吸收波段、近红外反射峰及短波红外反射峰等特征构建高光谱指数,从而对冬小麦产量进行预测。结果表明:可见光波段、近红外波段和短波红外波段的光谱反射率与产量从返青期到抽穗期分别达到显著负相关、显著正相关和显著负相关水平;通过分析光谱参量与产量的关系,由植被红边与近红外波段反射峰所定义的归一化植被指数(NDVI)与产量的统计相关特征在所有生育期都是极显著水平,统计相关性优于其他光谱参量,利用该参量所构建的非线性模型估产效果最好,可见利用NDVI指数进行产量预报效果更好。
Rapid and accurate estimation of grain crop production is of great significance to the state in formulating food policies and agricultural sustainable development. By using the advantages of surface hyperspectral remote sensing, we can obtain the fine spectrum of crop canopy and construct the hyperspectral index according to the features of green peak, red edge, water vapor absorption band, near-infrared reflection peak and short-wave infrared reflection peak, prediction. The results showed that the spectral reflectance in the visible, near-infrared and shortwave infrared bands was significantly negatively correlated with the yield from heading stage to heading stage, with a significant positive correlation and significant negative correlation. By analyzing the relationship between spectral parameters and yield, The statistical correlations of NDVI and yield, defined by vegetation red-fringe and near-infrared bands, were statistically significant at all growth stages and were statistically superior to other spectral parameters and were constructed using this parameter The non-linear model has the best estimation effect. It can be seen that it is better to use NDVI index to predict the yield.