Training Dataset for Deep Learning-Based Coastline Extraction of Sandy Beaches in Hainan Island

This dataset is a deep learning training set designed for accurate extraction of sandy beach coastlines in Hainan Island, with a focus on distinguishing between landward and seaward segments of the coastline. It supports the development and validation of deep learning models to automate coastline identification, providing reliable baseline data for quantifying coastline positions.In association with the submitted paper "Coastline variability and responses to climate forcing of a tropical island in the western Pacific marginal seas", this dataset serves as a foundational data source. The coastline extraction results derived from this training set underpin the paper's analyses of spatiotemporal variability in Hainan Island's sandy beach coastlines, and further support the quantification of coastline responses to climate forcing factors (e.g., sea-level rise, storm tracks) in the western Pacific marginal seas. For specific details, please refer to our literature materials "Deep learning - based automated coastline extraction from satellite imagery: A Sentinel-2 benchmark dataset approach".