基于低空无人机多光谱遥感的水稻倒伏监测研究  

Monitoring of lodged rice using low-altitude UAV based multispectral image

作　　者：田明璐 班松涛[2] 袁涛[1] 籍延宝 马超[1] 李琳一[1] TIAN Ming-lu;BAN Song-tao;YUAN Tao;JI Yan-bao;MA Chao;LI Lin-yi(Agricultural Information Institute of Science and Technology Shanghai Academy of Agricultural Science,Shanghai Engineer Research Center for Digital Agriculture,Shanghai 201403,China;College of Natural Resources and Environment,Northwest A & F University,Yangling 712100,China)

机构地区：[1]上海市农业科学院农业科技信息研究所,上海市数字农业工程技术研究中心,上海201403 [2]西北农林科技大学资源环境学院,杨凌712100

出　　处：《上海农业学报》2018年第6期88-93,共6页Acta Agriculturae Shanghai

基　　金：上海市农委项目[沪农科推字(2016)第1-5-10号];上海市科委项目(18391901100)

摘　　要：为了获取水稻倒伏信息,使用小型低空无人机搭载多光谱成像仪,对发生倒伏灾害的水稻田块进行遥感监测。结果表明:在多光谱影像上,倒伏水稻与正常水稻的光谱反射率在各个波段上均有差异,其中以绿波段和红边波段差异最为显著。在光谱分析的基础上,使用偏最小二乘判别分析法构建基于全波段光谱反射率的水稻正常/倒伏分类模型,取分类阈值k=0. 5,对多光谱影像上的正常水稻和倒伏水稻进行了有效区分,准确率分别达到98. 104%和99. 044%,精度能够满足水稻倒伏面积监测的需求。根据分类结果统计出研究区水稻倒伏面积为38 074. 4 m2,占总面积的73. 4%。研究结果表明,使用无人机多光谱遥感可以为水稻倒伏监测提供高效、精确的技术支持。The multispectral remote sensing sensor mounted on a low-altitude unmanned aerial vehicle(UAV) was used to monitor the lodged rice field in order to collect the lodging information. The results showed that the spectral reflectance of normal rice and lodged rice on the images were significantly different in every waveband,and the difference reached the maximum in green band and red-edge band. Based on the spectral analysis,the normal/lodged rice classification model was built by using partial least squares—discriminant analysis(PLS-DA) method on the basis of spectral reflectance of all bands. The classification threshold k was set as 0. 5. The normal rice and lodged rice were effectively distinguished using the PLS-DA model,and the classification accuracy reached 98. 104% and 99. 044%,respectively. The area of lodged rice field calculated from the classification map was 38 074. 436 m2,which was 73. 362% of the total rice field observed,indicating a severe damage. This paper proved that multispectral remote sensing based on UAV was an efficient tool to precisely monitor the rice lodging.

关 键 词：无人机 多光谱遥感 水稻 倒伏 偏最小二乘判别分析 

分 类 号：S511[农业科学—作物学] S252