Global 30m land-cover products with fine classification system in 2015
收藏地球大数据科学工程2019-09-29 更新2024-10-12 收录
下载链接：
https://data.casearth.cn/item/5d904b7a0887164a5c7fbfa0
下载链接
链接失效反馈官方服务：
资源简介：
Land cover information is indispensable and important in ecological environment assessment, national-condition monitoring, and macro-control analysis. Recently,owing to the rapidly increasing data-storage and computation capabilities, the needs of producing the global 30m land-cover products are increasingly urgent. Therefore, we proposed to combine the time-series MCD43A4 NBAR data and ESA CCI_LC land-cover products to develop a priori GSPECLib (Global Spatial-Temporal Spectral Library), which would replace the training samples in the traditional classification. Then, using the GSPECLib, time-series Landsat-8 surface reflectance and local adaptive random forest models, a novel global 30 m land-cover products with fine classification system in 2015 (GLC_FCS30-2015) was produced, which contained 30 fine land-cover types and achieved an overall accuracy of 81.4% and a kappa coefficient of 0.772
创建时间：
2022-03-30
