摘要
采用无人机影像进行绿地信息分类时,常利用影像光谱、纹理、形状等分类特征,忽视了通过无人机影像生成点云构建的数字表面模型(Digital surface model,DSM)和数字高程模型(Digital elevation model,DEM)差异特征。基于此,提出一种顾及无人机影像点云特征的绿地信息分类方法。方法首先基于摄影测量理论对研究区无人机影像进行空三计算,并生成点云,在此基础上构建DSM、DEM和数字正射影像(Digital Orthophoto Map,DOM);然后,利用DSM和DEM模型构建地物高度差异模型(normalized Digital Surface Model,nDSM);最后,利用可见光波段差异植被指数(Visible-band difference vegetation index,VDVI)对DOM进行植被与非植被分类,并结合nDSM对植被进行分类。以昆明市呈贡区白龙潭公园为研究区进行绿地信息分类,Kappa系数精度达到0.862,实验表明本文的方法对城市绿地调查具有实际意义。
When UAV images are used to classify vegetation information,classification features such as image spectrum,texture and shape are often used,while digital surface model(DSM)and digital elevation model(DEM)constructed by UAV image generation point cloud are ignored.A vegetation classification method considering the point cloud characteristics of UAV image is proposed in this paper.Firstly,based on the photogrammetry theory,the UAV images were calculated to generate image point cloud.DSM,DEM and Digital Orthophoto Map(DOM)were constructed.Then,the normalized Digital Surface Model(nDSM)is constructed by using DSM and DEM.Finally,the Visible band difference vegetation index(VDVI)is used to classify vegetation and non-vegetation in DOM.nDSM is used to distinguish different vegetation types.Taking Bailongtan Park in Chenggong District of Kunming City as the study area for green space information classification,the Kappa accuracy is 0.862.The experimental results show that the method of this paper is of practical significance to the investigation of urban green space.
作者
李佳
王明果
王云川
詹顺
段平
LI Jia;WANG Mingguo;WANG Yunchuang;ZHAN Shun;DUAN Ping(Department of Geography,Yunnan Normal University,Kunming 650500,China;Yunnan Institute of Geological Sciences,Kunming 650501,China;Geological Surveying and Mapping Institute of Yunnan Province,Kunming,Kunming 650218,China)
出处
《生态科学》
CSCD
2022年第5期11-18,共8页
Ecological Science
基金
国家自然基金项目(41961061)
云南省基础研究计划项目(2020FB059)
云南省教育厅科学研究基金(2021Y502)
云南师范大学2021年研究生科研创新基金(YJSJJ21-B82)。
关键词
绿地分类
无人机
可见光植被指数
点云
vegetation information classification
UAV
the Visible band difference vegetation index
point cloud
作者简介
李佳(1984-),女,湖北公安人,博士,副教授,博士生导师,主要从事无人机影像处理与应用研究,E-mail:keguigiser@163.com;通信作者:段平,男,博士,副教授,主要从事无人机影像处理与应用研究,E-mail:dpgiser@163.com。
