| 基于混合像元分解方法的康保县植被覆盖度估测 |
| |
| 引用本文： | 陈松,孙华,陈振雄,吴童.基于混合像元分解方法的康保县植被覆盖度估测[J].中南林业调查规划,2019(1):39-44. |
| |
| 作者姓名： | 陈松 孙华 陈振雄 吴童 |
| |
| 作者单位： | 中南林业科技大学林业遥感信息工程研究中心;林业遥感大数据与生态安全湖南省重点实验室;南方森林资源经营与监测国家林业局重点实验室;国家林业与草原局中南调查规划设计院 |
| |
| 基金项目： | 国家林业与草原局防治荒漠化管理中心荒漠化监测专项经费(101-9899);湖南省创新平台与人才计划(科技人才)项目(2015RS4048);中国博士后科学基金项目(2014M562147)资助 |
| |
| 摘 要： | 以康保县为研究区,使用Landsat8 OLI数据,结合134个野外样地调查数据,通过像元二分法、完全约束最小二乘法和随机森林三种方法来进行混合像元的分解,探讨混合像元分解方法对提取地表植被覆盖度的可行性。结果显示:基于随机森林的混合像元分解方法结果最优,F为0.664,RMSE为0.127;三种方法都能较为精确地估测出康保县的植被覆盖情况,且与实测数据较为拟合,因此采用混合像元分解的方法进行植被覆盖度估测是完全可行的。
|
| 关 键 词： | 植被覆盖度 混合像元分解 随机森林 完全约束最小二乘法 像元二分法 |
Estimation of Vegetation Coverage of Kangbao County Based on Spectral Unmixing Analysis Methods |
| |
| CHEN Song,SUN Hua,CHEN Zhenxiong,WU Tong.Estimation of Vegetation Coverage of Kangbao County Based on Spectral Unmixing Analysis Methods[J].Central South Forest Inventory and Planning,2019(1):39-44. |
| |
| Authors: | CHEN Song SUN Hua CHEN Zhenxiong WU Tong |
| |
| Institution: | (Research Center of Forestry Remote Sensing & Information Engineering Central South University & Technology,Changsha 410004,Hunan,China;Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security for Hunan Province,Changsha 410004,Hunan,China;Key Laboratory of State Forestry Administration on Forest Resources Management and Monitoring in Southern Area,Changsha 410004,Hunan,China;Central South Inventory and Planning Institute of National Forestry and Grassland Administration,Changsha 410014,Hunan,China) |
| |
| Abstract: | This paper focused on the vegetation coverage, using the Landsat8 OLI data for Kangbao County, Hebei Province. In this paper, 134 sample plots were systematically selected in Kangbao County and vegetation coverage data were collected. In addition, the pixel binary method, the completely constrained least square method and the random forest method were conducted to spectral unmixing analysis to explore the feasibility of extracting vegetation coverage by spectral unmixing analysis. The results showed that: the spectral unmixing analysis method based on random forest model was the best, with R2 coefficient 0. 664 and RMSE coefficient 0. 127. The three methods could all accurately estimate the vegetation coverage in Kangbao County and the estimations were close to the actual data. Therefore, it was feasible to estimate the vegetation coverage by the method of the spectral unmixing analysis. |
| |
| Keywords: | vegetation coverage spectral unmixing analysis random forest fully constrained least squares pixel dichotomy |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
