基于车载式冠层光谱传感器的玉米拔节期叶绿素含量诊断
Estimation of maize chlorophyll content by vehicle-mounted crop canopy sensor
-
摘要: CropspecTM是一种基于735 nm 和808 nm的车载式主动作物冠层光谱传感器,能够快速、无损地检测作物氮素营养状态。为了评价其检测精度,针对农大8号和京农科等2种玉米作物品种,使用该检测系统在拔节期采集作物冠层在808nm和735nm波段处的反射率。然后组合计算了DVI735, NDVI735, PVI735和 RDV735 等常规的植被指数,并基于RVI735构造了一种新的植被指数MRVI735。通过分析各植被指数与叶绿素含量指标SPAD值之间的相关关系得出 :对于农大8号,MRVI735、NDVI735和RVI735与叶绿素含量指标的相关性较好,相关系数分别是:-0.7482、-0.6763和-0.6786,达到强相关水平。对于京农科,NDVI735、MRVI735和RVI735与叶绿素含量指标的相关性较好,相关系数分别是:0.7270、0.7252和0.7245,达到强相关水平。对于2个玉米品种,都分别选取了相关系数最好的一个和两个植被指数为参数,分别建立了一元线性回归模型和二元线性回归模型。农大8号的一元模型和二元模型的R2Abstract: Crop canopy sensor based estimation is an efficient approach to estimate chlorophyll content of crop.CropspecTM is an active vehicle mounted crop canopy sensor, with the fixed wavebands being 735 nm and 808 nm.Using pulsed laser diodes as light source, the sensor is designed to irradiate the crop at an oblique view in height up to 4 meters for measuring crop canopy of various crops accurately in a large footprint.In this paper, a set of vegetation indices(DVI735, RVI735, NDVI735, PVI735, RDV735 and MRVI735) were calculated and measured by the CropspecTM sensor, and the correlation analysis between vegetation indices and chlorophyll content index were explored.The results showed that MRVI735 and NDVI735 had high correlation with coefficients R of -0.7842 and 0.7240 corresponding to ND8 and JNK respectively, proving that RVI735 was more appropriate parameter for chlorophyll content estimation.A ULR model and a MLR model were established for each cultivar.For ND8, R2
下载: