基于无人机多光谱遥感的苎麻叶绿素含量反演

Inversion of Chlorophyll Content in Ramie Based on UAV Multispectral Remote Sensing

苎麻叶绿素含量与其光合作用及有机质积累能力有着极大的相关性,对监测其生长、衰老有重要意义。无人机遥感技术的发展为快速、无损、高效监测作物叶绿素提供了新方法,但利用该技术对苎麻叶绿素含量进行监测的研究还未见报道。本研究基于多时序获取的无人机多光谱遥感影像,通过分析遥感指数与苎麻叶片叶绿素含量之间的相关性,采用多种机器学习方法构建了不同生育期叶绿素含量反演模型。结果表明:基于无人机多光谱遥感反演苎麻叶绿素含量的最佳反演模型为成熟期的随机森林回归模型,决定系数为0.892,均方根误差为0.116。本研究结果可为利用无人机多光谱遥感预测苎麻叶绿素含量提供方法参考。

The chlorophyll content of ramie has great correlation with its photosynthesis and organic mat-ter accumulation capacity,which is important for monitoring its growth and aging.The development of UAV re-mote sensing technology provides a new method for rapid,nondestructive and efficient monitoring of crop chlo-rophyll,but no research has been reported on the use of this technology for monitoring chlorophyll content in ramie.In this study,based on the UAV multispectral remote sensing images acquired in multiple time series,we constructed an inverse model of chlorophyll content at different fertility stages using various machine learn-ing methods by analyzing the correlation between remote sensing indices and chlorophyll content of ramie leav-es.The results showed that the best inversion model for ramie leaf chlorophyll content based on UAV multi-spectral remote sensing was random forest regression model at maturity with determination coefficient as 0.892 and root mean square error as 0.116.The results of this study provided a method reference for predicting the chlorophyll content of ramie using UAV multispectral remote sensing.