UGS-1m

The project provides 1-meter UGS maps of 31 major cities in China (UGS-1m), which were generated by a deep learning (DL) framework. A UGSet and a UGSNet are included in the DL framework for large-scale and high-resolution UGS mapping.

UGS-1m: a fine-grained UGS map product of 31 major cities in China of 1 meter
UGSet: a large benchmark dataset to support and foster the UGS research
UGSNet: a fully convolutional network for fine-grained UGS mapping

diagram

News

We are continuing to update our UGS products in other regions of China through the proposed method, which will grouped by province/municipality/autonomous region! The updating data can be accessed from OneDrive for the time being.

2023-05-15: The UGS data of Beijing, Tianjin, Shanghai, Chongqing, Taiwan, Hong Kong and Macau ara now available.

Reference

This work has been published in ESSD [Paper Link].

@Article{essd-15-555-2023,
AUTHOR = {Shi, Q. and Liu, M. and Marinoni, A. and Liu, X.},
TITLE = {UGS-1m: fine-grained urban green space mapping of 31 major cities in China based on the deep learning framework},
JOURNAL = {Earth System Science Data},
VOLUME = {15},
YEAR = {2023},
NUMBER = {2},
PAGES = {555--577},
URL = {https://essd.copernicus.org/articles/15/555/2023/},
DOI = {10.5194/essd-15-555-2023}
}

UGS-1m product

The UGS-1m product provides the fine-grained UGS maps of 31 major cities in China, which is generated based on a deep learning (DL) framework.

The product is now available at ScienceDB. The Google Earth imagery used is available at OneDrive

@misc{qian2023ugs1m,
  author       = {Qian Shi and Mengxi Liu and Andrea Marinoni and Xiaoping Liu},
  title        = {UGS-1m: Fine-grained urban green space mapping of 31 major cities in China based on the deep learning framework},
  year         = 2023,
  month        = jan,
  publisher    = {Science Data Bank},
  version      = {V1},
  doi          = {10.57760/sciencedb.07049},
  url          = https://doi.org/10.57760/sciencedb.07049
}

UGS-1m

UGSet

A largescale high-resolution urban green space dataset (UGSet). The dataset is now available at ScienceDB and Onedrive. UGSet

UGSNet

The model code is now available at Github. framework UGSNet

Contact

Correspondence: liumx23@mail2.sysu.edu.cn