冬小麦条锈病严重度不同估算方法对比研究

为了提高遥感监测小麦条锈病病害严重度的准确性，寻找小麦病害的较优反演模型，在国家精准农业示范研究基地基于野外定位调查小麦病情指数及冠层光谱数据，利用与小麦病害发生呈显著关系且有效反映植被生理生长状况的7种高光谱植被指数，尝试分别采用PLS(偏最小二乘回归)、BP神经网络和植被指数经验法三种方法建立小麦条锈病病情反演模型，并进行比较分析。结果表明：三种方法病害严重度预测值与实测值间的R2分别为0.936，0.918，0.767。采用偏最小二乘回归方法监测小麦病情指数效果更好，为探寻不同种类植被指数对模型的贡献，尝试用代表植被绿度的NDVI, GNDVI, MSR和代表水分含量的NDWI和MSI植被指数分别作为PLS模型的输入变量，建立病害反演模型。结果表明：小麦条锈病中，叶片叶绿素含量的变化比冠层水含量的变化对病情指数更为敏感，对病害有更好的解释作用。然而，两模型精度都低于七种植被指数全部参与时的预测结果，即输入变量中采用多种植被指数比用单类指数模拟准确度高。

Comparative Research on Estimating the Severity of Yellow Rust in Winter Wheat

In order to improve the accuracy of wheat yellow rust disease severity using remote sensing and to find the optimum inversion model of wheat diseases, the canopy reflectance and disease index (DI) of winter wheat under different severity stripe rust were acquired. The three models of PLS (Partial Least Square), BP neural network using seven hyperspectral vegetation indices which have significant relationship with the occurrence of disease and vegetation index (PRI) were adopted to build a feasible regression model for detecting the disease severity. The results showed that PLS performed much better. The inversion accuracy of PLS method is best than of the VI (PRI, Photochemical Reflectance Index) and BP neural network models. The coefficients of determination (R2) of three methods to estimate disease severity between predicted and measured values are 0.936，0.918 and 0.767 respectively. Evaluation was made between the estimated DI and the measured DI, indicating that the model based on PLS is suitable for monitoring wheat disease. In addition, to explore the different contributions of diverse types of vegetation index to the models, the paper attempts to use NDVI, GNDVI and MSR which on behalf of vegetation greenness and NDWI and MSI that represents the moisture content to be input variables of PLS model. The results showed that, for the wheat yellow rust disease, changes in chlorophyll content is more sensitive to the disease severity than the changes in water content of the canopy . However, the accuracy of the two models are both lower than predicted when participating in all seven vegetation indices, namely using several species of vegetation indices tends to be more accurate than that using single category. It indicated that it has great potential for evaluating wheat disease severity by using hyper-spectral remote sensing.