用高光谱微分指数监测冬小麦病害的研究

工田间诱发不同等级小麦条锈病，在不同生育期测定染病冬小麦冠层光谱、生理生化参数以及相应的病情指数。 对小麦冠层一阶微分光谱进行分析，结果表明随病情指数增大，一阶微分光谱在绿边(500～560 nm)内逐渐增大，在红边(680～760 nm)内逐渐降低。红边核心区(725～735 nm)内一阶微分总和(SDr′)与绿边核心区(520～530 nm)内一阶微分总和(SDg′)的比值，与病情指数具有极显著线性负相关性，相关系数r2＝0.921(n＝28)，且能够在症状出现前12 d识别出健康作物与病害作物。因此，微分植被指数SDr′/SDg′能够监测并反演作物病害信息。研究结果对利用高光谱遥感获取作物病害信息具有实际应用价值，对提高粮食产量、保证粮食安全具有重要意义。

Using Hyperspectral Derivative Index to Monitor Winter Wheat Disease

The canopy reflectance of winter wheat that infected different severity stripe rust was measured through artificial inoculation, the disease index (DI) of the wheat corresponding to the spectra was acquired in the field, and the parameters of biochemistry and biophysics were measured indoors. The 1st derivatives were analyzed. The results show that the 1st derivative values increase at the green edge (500-560 nm), while decrease at the red edge (680-760 nm) with DI increasing. The ratio of the sum of derivatives within the red edge (SDr′) to the sum of derivatives within the green edge (SDg′) has a higher negative linear correlation with DI, with a coefficient of determination r2=0.921 0(n=28), and that can be use to identify the healthy and disease crops 12 days before symptoms appearing. Therefore, the derivative vegetation index SDr/SDg can be used to monitor crops disease information. The conclusion is significant and may find application in acquiring crops disease information using hyperspectral remote sensing, and has a important meaning for increasing yields of crops and ensuring security of food supplies.