基于无人机遥感监测滩涂湿地入侵种互花米草植被覆盖度  

Fractional vegetation cover of invasive Spartina alterniflora in coastal wetland using unmanned aerial vehicle (UAV) remote sensing

作　　者：周在明[1] 杨燕明[1] 陈本清[1] 

机构地区：[1]国家海洋局第三海洋研究所,福建厦门361005

出　　处：《应用生态学报》2016年第12期3920-3926,共7页Chinese Journal of Applied Ecology

基　　金：福建省自然科学基金项目(2015J05085);促进海峡两岸科技合作联合基金项目(U1405234);国家海洋局第三海洋研究所基本科研业务费项目(HE150805-14B)资助~~

摘　　要：为了实现对沿海滩涂湿地资源与生态环境的有效管理与利用,需要对区域内的入侵种互花米草进行高精度的监测与分析.本文以福建三沙湾为试验区,以低空无人机获取的可见光和多光谱影像为数据源,对互花米草植被覆盖度进行监测与分析,通过NDVI指数模型获取了多光谱影像的植被覆盖度信息,以可见光影像为参考进行了精度检验.结果表明:影像区互花米草植被覆盖度以40%~60%和60%~80%的中高和高等级覆盖度为主.NDVI模型估算值与真实值之间的均方根误差RMSE为0.06,决定系数R^2为0.92,两者具有较好的一致性.The effective management and utilization of resources and ecological environment of coastal wetland require investigation and analysis in high precision of the fractional vegetation cover of invasive species Spartina alterniflora. In this study, Sansha Bay was selected as the experimental region, and visible and multispectral images obtained by lowaltitude UAV in the region were used to monitor the fractional vegetation cover of S. alterniflora. Fractional vegetation cover parameters in the multispectral images were then estimated by NDVI index model, and the accuracy was tested against visible images as references. Results showed that vegetation covers of S. alterniflora in the image area were mainly at medium high level （40%-60%） and high level （60%-80%）. Root mean square error （RMSE） between the NDVI model estimation values and true values was 0.06, while the determination coefficient R^2 was 0.92, indicating a good consistency between the estimation value and the true value.

关 键 词：互花米草 植被覆盖度 无人机 可见光影像 多光谱影像 

分 类 号：S45[农业科学—植物保护]