摘要
高效、无损的监测作物长势是现代精准农业的核心环节,无人机平台因具有成本低、数据获取效率高、测试高度及测试时间可按需调节等优点,在监测作物长势中发挥着地面平台和高空平台无法比拟的优势。本研究以小麦为研究对象,应用无人机搭载RedEdge-M多光谱相机获取主要生育时期的小麦冠层多光谱影像,并同步取样测量小麦叶片SPAD、地上部鲜重和干重,进一步探索基于无人机平台获取多光谱影像的预处理方法,提取小麦冠层反射率并筛选出适合作物长势监测的植被指数,构建基于无人机平台的小麦长势监测模型,结果表明,基于NDVI、SAVI、CCCI构建的多元线性回归模型精度更高、稳定性更好。预测小麦SPAD值的最佳模型为y=19.765+7.522NDVI+18.362SAVI+25.629CCCI,R^(2)为0.965;预测小麦地上部干重的最佳模型为y=-0.508+0.603NDVI+0.325SAVI+0.032CCCI,R^(2)为0.951;预测小麦地上部鲜重的最佳模型为y=-2.217+2.923NDVI+2.213SAVI-1.417CCCI,R^(2)为0.766。本研究结果可为园区和农场尺度小麦长势的实时监测提供有效技术支撑。
Efficient and non-destructive monitoring of crop growth is the core of modern precision agriculture. UAV platform has the advantages of low cost, high efficiency of data acquisition, adjustable test height and test time on demand, and plays an unparalleled advantage compared with ground platform and high-altitude platform in monitoring crop growth. Taking wheat as the research object, the RedEdge-M multispectral camera was used to obtain the multispectral images of wheat canopy in the main growth periods, and simultaneously sampled and tested SPAD and fresh and dry weights of aboveground parts, and then further explored the preprocessing method of obtaining multispectral images based on UAV platform, extracted the wheat canopy reflectance, and screened out the vegetation indexes suitable for crop growth monitoring to construct wheat growth monitoring model based on UAV platform. The results showed that the MLR model based on NDVI, SAVI and CCCI was more accurate and stable. The best monitoring model for predicting wheat SPAD value was y=19.765+7.522 NDVI+18.362 SAVI+25.629 CCCI with the R2 as 0.965. The best monitoring model for predicting dry weight content of aboveground part was y=-0.508+0.603 NDVI+0.325 SAVI+0.032 CCCI with the R2 as 0.951. The best model for predicting fresh weight content of aboveground part was y=-2.217+2.923 NDVI+2.213 SAVI-1.417 CCCI with the R2 as 0.766. This study could provide an effective technical support for the real-time monitoring of wheat growth at the park and farm.
作者
牛鲁燕
蒋风伟
张俊丽
孙家波
张晓艳
卢德成
刘延忠
Niu Luyan;Jiang Fengwei;Zhang Junli;Sun Jiabo;Zhang Xiaoyan;Lu Decheng;Liu Yanzhong(Institute of Science and Technology Information,Shandong Academy of Agricultural Sciences,Jinan 250100,China;People’s Government of Xingcun Town,Sishui County,Sishui 273209,China;Juancheng Bureau of Agricultural and Rural Affairs,Juancheng 274600,China;Shandong Cotton Research Center,Jinan 250100,China)
出处
《山东农业科学》
北大核心
2021年第3期139-145,共7页
Shandong Agricultural Sciences
基金
国家重点研发计划项目(2017YFD0301004-3)。
关键词
无人机
多光谱影像
小麦
长势监测
UAV
Multispectral imagery
Wheat
Growth monitoring
