This dataset is the forest type cover data with 10m spatial resolution of China (2018). Combining Landsat and sentinel-2 remote sensing images as data sources, the spectral and spatiotemporal feature sets of different forest types have been established. Using Landsat and time series harmonic analysis to establish a time feature set. Based on the spectral-temporal feature set, supported by reference data, the random forest recursive feature elimination algorithm is used to study and establish the main features of different regions. According to the spectral-spatial-temporal feature sets of different regions, four machine learning algorithms are used to establish a forest type classification model. Then, using the determined best-fitting model, a forest type map with a spatial resolution of 10 in Southeast China in 2018 was generated. The data format is TIF, the spatial range is Southeast China, and the time is 2018.
Comment list ( 0 )