欢迎访问《麦类作物学报》编辑部网站

作者	单位
范友波，顾晓鹤，王双亭，杨贵军，王磊，王立志，陈召霞	（1.河南理工大学测绘与国土信息工程学院，河南焦作 454000； 2.国家农业信息化工程技术研究中心，北京 100097）

摘要点击次数: 984

全文下载次数: 688

中文摘要:

为探讨利用高光谱技术快速无损地监测小麦白粉病灾情的方法，通过人工田间诱发白粉病，在灌浆期对不同发病等级（病情指数）的冬小麦进行冠层高光谱测定，对原始光谱数据进行一阶微分处理，筛选最佳光谱特征参量和植被指数，构建冬小麦白粉病病情指数反演模型。结果表明，在冠层尺度，小麦白粉病“红边”位置均在730 nm左右（±1 nm）；经验证，5种模型中三角植被指数（TVI）模型估算精度最好，R和RMSE分别达到了0.700和0.112，与精度最低的优化土壤调节植被指数(OSAVI)模型相比，R提高了0.071，RMSE降低了0.013。小麦白粉病“红边”蓝移现象并不明显；五种模型R都达到了0.6以上，说明高光谱技术都能够有效地对冬小麦白粉病病情指数进行无损、快速、精确的反演，其中TVI的反演精度最佳。

英文摘要:

In order to enable rapid and non-destructive monitor of winter wheat powdery mildew utilizing remote sensing technologies, we artificially introduced the disease and measured the spectrums of different disease degrees (quantified in disease index) in filling stage, then performed the first-order differential transformation procedure to the primary spectrums and built the model for estimating the disease index of winter wheat.Results showed that the red edge position was mainly at 730±1 nm. The results of precision verification indicated that the triangle vegetation index (TVI) model amongst the five kinds of model gained a better precision of estimating, of which R and RMSE could reach 0.700 and 0.112, respectively. When compared with the lowest accuracy of the inversion model with optimized soil-adjusted vegetation index (OSAVI), R was increased by 0.071 and RMSE was reduced by 0.013. Results showed that the red edge position shifted to the shorter wavelength was not obvious at canopy level. The R values of all five models reached above 0.6, which indicated that hyperspectral technology can effectively, non-destructively, rapidly, accurately retrieve winter wheat powdery mildew disease index (DI). The TVI could be considered as the optimal parameter used to inverse the disease index of powdery mildew of winter wheat.

查看全文查看/发表评论下载PDF阅读器

关闭