Plot-Rice v1.0: A global plot-based rice benchmark dataset with spatiotemporal heterogeneity for scientific deep learning

This dataset (Plot-Rice v1.0) offers a global rice benchmark dataset for scientific deep learning at a 10-meter resolution for the year 2023. Plot-Rice v1.0 is constructed based on Sentinel-1 and Sentinel-2 images, encompassing plot-level rice labels and corresponding multi-source feature time series from 20 countries worldwide. It fully considers the spatiotemporal heterogeneity of rice and supports continuous updates, providing a data benchmark for performance comparisons of deep learning models.