基于低空无人机遥感的高粱生长状况监测  

Monitoring of sorghum growth based on UAV remote sensing

作　　者：陈虹 谢玲[2] 陈林琳 Chen Hong;Xie Ling;Chen Linlin(Nantong College of Science and Technology,Nantong,226001,China;Nanjing University of Science and Technology ZiJin College,Nanjing,210023,China)

机构地区：[1]南通科技职业学院,江苏南通226001 [2]南京理工大学紫金学院,南京市210023

出　　处：《中国农机化学报》2021年第4期170-175,共6页Journal of Chinese Agricultural Mechanization

基　　金：南通市科技项目(JC2018135)。

摘　　要：配备多光谱相机的无人机可实现对农作物生长状况的快速无损监测,为评估无人机遥感监测高粱作物长势的可行性和准确性,利用无人机搭载的多光谱相机获取高粱拔节期、抽穗开花期、灌浆成熟期多光谱遥感图像,构建常用的4种植被指数与叶面积指数LAI和植被覆盖度FVC之间的回归模型。经过精确度评价,确定归一化差异植被指数NDVI为最优植被指数,LAI-NDVI和FVC-NDVI估算模型的决定系数R^(2)分别为0.91和0.88,均方根误差RMSE分别为0.28和0.06;平均绝对误差MAPE分别为11%和8%。基于此,选择归一化差异植被指数NDVI,分析LAI和FVC无人机遥感估算值和实测值之间的关系,通过交叉验证得到LAI值:R^(2)=0.94,RMSE=0.16,MAPE=13%;FVC值:R^(2)=0.90,RMSE=0.05,MAPE=4%,说明两者存在高度拟合性。结果表明:根据无人机遥感得到的归一化差异植被指数NDVI可准确地估算高粱作物的叶面积指数和植被覆盖度,无人机遥感适用于对高粱作物生长状态的监测。In order to evaluate the feasibility and accuracy of UAV remote sensing monitoring of sorghum crop growth,the multispectral remote sensing images of sorghum jointing stage,heading and flowering stage,and grouting maturity stage were obtained by using the multi spectral camera carried by UAV.Regression models were then constucted between 4 commonly used vegetation index and leaf area index(LAI)and the fractional vegetation cover(FVC).After the accuracy evaluation,NDVI was determined as the optimal vegetation index.The determination coefficients R^(2) of LAI-NDVI and FVC-NDVI estimation models were 0.91 and 0.88,the root mean square error(RMSE)was 0.28 and 0.06,and the mean absolute percentage error(MAPE)was 11%and 8%.Based on this,the normalized difference vegetation index(NDVI)was selected to analyze the relationship between the remote sensing estimated value and the measured value of LAI and FVC.Through cross validation,the LAI value was R^(2)=0.94,RMSE=0.16,MAPE=13%;FVC value:R^(2)=0.90,RMSE=0.05,MAPE=4%,indicating that there is a high degree of fitting between them.The results showed that the NDVI obtained by UAV remote sensing could accurately estimate the leaf area index and fractional vegetation cover of sorghum crops,and UAV remote sensing was suitable for monitoring the growth status of sorghum crops.

关 键 词：无人机遥感 高粱 生长状况 归一化差异植被指数 叶面积指数 植被覆盖度 

分 类 号：S252[农业科学—农业机械化工程]