Remotely Sensed Paddy Rice Map of South Korea (2017-2021)

This dataset includes paddy rice maps in South Korea from 2017 to 2021 with 10 m resolution, which was produced by analyzing time-series Sentinel-1 images with recurrent U-Net deep learning architecture. The paddy rice maps are a product of deep learning model predictions and DO NOT represent ground truth information.

The detailed algorithm for producing the dataset can be found in the following paper: https://doi.org/10.1080/15481603.2023.2206539

The used modeling architecture is indicated by "RU-net 3" in the paper, and the consisting products in the dataset is as follows:

PR_in_[Year] : Probabiltiy of paddy rice cultivation area inside the paddy boundary (levee)

PR_bd_[Year] : Probabiltiy of levee

PR_in and PR_bd files have a scale factor vaule of 1,000,000.

PR_bi_[Year] : Binary map of paddy rice, Due to the pixel size much larger than the levee width, the pixels labeled as levees inevitably include a large portion of the cultivation area. Therefore, binary map represent the sum of PR_in and PR_bd exceeding 0.5 threshold.

Please cite the paper when using this dataset.

This work was supported by the International Research and Development Program of the National Research Foundation of Korea (NRF), funded by the Ministry of Science and ICT [2021K1A3A1A78097879], and partially supported by the European Commission under contract H2020-CALLISTO [101004152]