摘要
基于高分光学遥感影像技术,对湖南省油茶产业大县汉寿县油茶林地进行分类特征提取、标注样本类别及建立分类模型等操作,实现研究区油茶林地高精度识别。结果显示:利用基于高分光学遥感影像的深度网络智能化提取油茶林地及人工确认的数据分析方法能够很好识别汉寿县油茶林地的空间分布,与2022年全县油茶精细化调查面积数据对比后精度较高(优于70%);空间分布上,汉寿县太子庙镇的油茶分布集中且面积最大,和地面调查的面积数据相比,该镇的油茶分布面积提取精度最高,为92.57%。研究方法能够较为精准地识别汉寿县油茶林地斑块,从而为湖南省油茶林地遥感普查与保护规划提供重要参考。
Based on high-resolution optical remote sensing image technology,classification feature extraction,annotated sample category,and establishment of classification models were carried out on the Camellia Oleifera forest land in Hanshou County,a major Camellia Oleifera industry county in Hunan Province,to achieve highprecision recognition of the Camellia Oleifera forest land in the research area.The results show:the data analysis method of deep network intelligent extraction and manual confirmation of Camellia oleifera forest land based on high-resolution optical remote sensing images can effectively identify the spatial distribution of Camellia oleifera forest land in Hanshou County,and compared with the refined survey area data of Camellia oleifera in the whole county in 2022,the accuracy is relatively high(better than 70%);in terms of spatial distribution,the distribution of Camellia oleifera in Taizimiao Town,Hanshou County,is relatively concentrated and largest in area,and compared with the field survey area data,the extraction accuracy of Camellia oleifera distribution area is the highest at 92.57% in Taizimiao Town.The research method can accurately identify the patches of Camellia oleifera forest in Hanshou County,therefore,it can also provide important reference for remote sensing survey and protection planning of Camellia oleifera forest in Hunan Province.
作者
杨文军
张杨
王福生
瞿跃辉
YANG Wenjun;ZHANG Yang;WANG Fusheng;QU Yuehui(Hunan Prospecting Designing&Research General Institute for Agriculture Forestry&Industry,Changsha 410007,Hunan,China;Central South University of Forestry&Technology,Changsha 410004,Hunan,China)
出处
《中南林业调查规划》
2023年第4期30-34,62,共6页
Central South Forest Inventory and Planning
基金
湖南省科学技术厅“高新技术产业科技创新引领计划(科技攻关类)——空天高分遥感大数据智能服务及森林资源监测应用技术”(2020GK2039)。
关键词
光学遥感
识别技术
油茶
林地
湖南
optical remote sensing
identification technology
Camellia oleifera
forest land
Hunan
