This dataset contains vectorized rooftop data of buildings in 90 Chinese cities (selected based on administrative hierarchy and regional distribution; see Attachment 1 for the city list). It was primarily developed using deep learning semantic segmentation models and multi-source remote sensing imagery. The workflow involves: 1) preprocessing raw imagery, followed by stratified sampling and visual interpretation based on city hierarchies and regional distribution to create training and testing datasets; 2) training a deep learning semantic segmentation model with the prepared data to optimize it for rooftop extraction tasks, with model performance evaluated using standard metrics in deep learning; and 3) applying the trained model to automatically extract and vectorize building rooftops across the 90 cities. This dataset provides critical foundational data for urban and national-scale research related to building rooftops, such as solar energy potential assessments and urban planning.
With authorization from the Smart City Sensing and Simulation Laboratory at Nanjing Normal University, CnOpenData has included this dataset in its public data repository for academic access. A sample dataset is available below, while the full dataset can be downloaded via the National Tibetan Plateau Data Center (http://data.tpdc.ac.cn/zh-hans/news/c1b0ee71-2ceb-4276-8f5c-b53d57caa88d).
Dataset Description
-
Dataset Year:
The dataset utilizes deep learning semantic segmentation models and multi-source remote sensing imagery for automated rooftop identification and extraction. The original imagery was acquired in 2020, with extracted rooftop raster pixels at 1m resolution. The actual year of identified rooftop vector features depends on the acquisition time of the remote sensing imagery. -
City Hierarchy:
The 90 cities are classified into four administrative tiers: sub-national level (副国级), provincial level (正省级), sub-provincial level (副省级), quasi-sub-provincial level (准副省级), and prefectural level (正厅级). For details, refer to the data download section.
Technical Specifications
- Scale: None
- Projection: Albers
- File Size: 143,360.0 MB
- Data Format: Esri Shapefile
- Spatial Extent: North: 53.55°; South: 3.86°; West: 73.66°; East: 135.05°
Field Descriptions
Sample Data
Beijing (Original file in Esri Shapefile format; table below shows a preview of the attribute data)
Citation
-
Dataset Citation:
Smart City Sensing and Simulation Laboratory, Nanjing Normal University. (2021). Vectorized rooftop area data for 90 cities in China (2020). National Tibetan Plateau Data Center. DOI:10.11888/Geogra.tpdc.271702, CSTR:18406.11.Geogra.tpdc.271702. -
Publication Citation:
Zhang, Z., Qian, Z., Zhong, T., et al. (2022). Vectorized rooftop area data for 90 cities in China. Scientific Data, 9(1), 1–12. https://doi.org/10.1038/s41597-022-01550-9
Update Frequency
Irregular updates