基于无人机多光谱图像监测棉花冠层含水量

Cotton Canopy Moisture Monitoring Based on Unmanned Aerial Vehicle Multispectral Image

使用无人机搭载的多光谱相机获取田间遥感影像,通过相关性计算选取合适的波段组合,基于多光谱影像间的波段运算得到植被指数(VIs),采用最小二乘法构建棉花冠层含水量反演模型。结果表明,红波段(680nm)和近红外1波段(800nm)间的光谱特征与棉花冠层含水量相关性最高,由此光谱区间构建了归一化植被指数(NDVI)和比值植被指数(RVI),基于NDVI的二阶多项式回归得到了较好的预测结果,R^(2)在0.69以上。使用此方法可以实现棉花冠层含水量的快速、无损监测,从而为田间精准灌溉提供技术支持。

In this study,the multi-spectral camera carried by the UAV was used to obtain remote sensing images of the field,and then the appropriate band combination was selected through correlation calculation,and the vegetation index(VIs)was obtained based on the band calculation between the multi-spectral images.Finally,the least square method was used to construct cotton Canopy Water Content Inversion Model.This study shows that the spectral characteristics between the red band(680nm)and near-infrared band 1(800nm)have the highest correlation with the cotton canopy water content.From this spectral interval,the normalized vegetation index(NDVI)and the ratio vegetation index(RVI)are constructed.The second-order polynomial regression based on NDVI has obtained better prediction results,and R^(2) is above 0.69.The method used in this study has realized the rapid and non-destructive monitoring of the cotton canopy water content,which can provide technical support for precise field irrigation.