Datasets

👉 30 m China Multi-Temporal Built-up Height (CMTBH-30) Dataset

The CMTBH-30 was developed using the MTBH-Net deep learning network, incorporating GEDI L2A V2 as height labels and Landsat and PALSAR as input features.

📥 Download: https://data-starcloud.pcl.ac.cn/iearthdata/
📖 Reference: Chen P, Huang H, Qin P, et al. Characterizing dynamics of building height in China from 2005 to 2020 based on GEDI, Landsat, and PALSAR data. Remote Sensing of Environment, 2025, 325: 114776. DOI: 10.1016/j.rse.2025.114776.

👉 Building Roof Dataset for Typical Chinese Cities

The building roof semantic segmentation dataset was created through manual digitalization, characterized by a spatial resolution of 0.65 m and covering an area of 573.17 km². The dataset includes 5,175 pairs of 512 × 512 image tiles with red, green, blue, and near-infrared bands.

📥 Download: https://data-starcloud.pcl.ac.cn/iearthdata/
📖 Reference: Chen P, Huang H, Ye F, et al. A benchmark GaoFen-7 dataset for building extraction from satellite images. Scientific Data, 2024, 11(1): 187. DOI: 10.1038/s41597-024-03009-5.