摘要
本文以芜湖市繁昌县为研究对象,基于Sentinel-2A数据与Landsat 8数据,利用像元二分法、线性光谱混合模型对研究区进行植被覆盖度估算,并对试验结果进行分析,探究两种数据在植被覆盖度反演方面的能力差异,比较了不同方法在研究区反演中的优劣性。试验结果表明,繁昌县植被覆盖状况较好;利用Landsat 8数据和Sentinel-2A数据分析得出的植被覆盖度具有较强的一致性,并且当利用线性光谱混合模型反演植被覆盖度时,Sentinel-2A数据总体精度比Landsat 8数据总体精度高3.84%;利用像元二分法反演植被覆盖度时,Sentinel-2A数据总体精度比Landsat 8数据总体精度高2.725%。总体而言,Sentinel-2A的植被覆盖度反演能力优于Landsat 8;像元二分法能较好地适用于研究区的植被覆盖度反演。
In this paper,Fanchang county of Wuhu city is taken as the research object.Based on sentinel-2 A data and landsat 8 data,the pixel dichotomy,linear spectral mixing model and three band gradient difference method are used to estimate the vegetation coverage in the study area.The experimental results are analyzed to explore the ability difference of the two data in vegetation coverage inversion,The advantages and disadvantages of different inversion methods in the study area are compared.The results shows that the vegetation coverage in Fanchang county is good;The results show that the vegetation coverage obtained from the analysis of Landsat 8 data and Sentinel-2 A data has a strong consistency,and the overall accuracy of Sentinel-2 A data is 3.84%higher than that of Landsat 8 data when the linear spectral mixture model is used to retrieve the vegetation coverage,and the overall accuracy of Sentinel-2 A data is 2.725%higher than that of Landsat 8 data when the pixel dichotomy method is used to retrieve the vegetation coverage.In general,Sentinel-2 A has better retrieval ability than Landsat 8;The pixel dichotomy method is suitable for vegetation coverage inversion in the study area.
作者
吴青云
高飞
李振轩
车子杰
WU Qingyun;GAO Fei;LI Zhenxuan;CHE Zijie(College of Civil Engineering,Hefei University of Technology,Hefei 230000,China)
出处
《测绘通报》
CSCD
北大核心
2021年第S01期104-108,113,共6页
Bulletin of Surveying and Mapping
基金
2018年安徽省测绘科技专项资金(CHZX201801)
中央高校基本科研业务费专项资金(JZ2021HGTA0167)
