A dataset of winter crops map in Henan Province from 2019 to 2021

First, the Landsat 7 / 8 and sentinel-2a / b remote sensing images are fused to construct a high spatial-temporal resolution time series data set. Secondly, the phenological differences of overwintering crops and other vegetation under different phenological periods are analyzed, and phenological indicators for classification are extracted, including start of season (SOS), start date of peak (SDP), end of season (EOS), green up speed (Gus) and growing season length (GSL).

Finally, based on these phenological indexes, a decision tree model is constructed to extract winter crops at the pixel scale. The phenological algorithm constructed in this study can accurately extract the planting area of overwintering crops, and its user, producer, overall accuracy and kappa coefficient are 98.00%, 98.36%, 97.77% and 0.94 respectively. This data set can provide basic data for the growth monitoring and yield prediction of overwintering crops, help decision makers and producers to formulate reasonable policies and risk management strategies, and also provide data reference for scientific researchers in relevant fields.