CBRA: The first multi-annual (2016-2021) and high-resolution (2.5 m) building rooftop area dataset in China derived with Super-resolution Segmentation from Sentinel-2 imagery

Large-scale and up-to-date maps of building rooftop area (BRA) are crucial for addressing policy decisions and sustainable development. In addition, as a fine-grained indicator of human activities, BRA could contribute to urban planning and energy modeling to provide benefits to human well-being. However, existing large-scale BRA datasets, such as those from Microsoft and Google, do not include China, hence there are no full-coverage maps of BRA in China. To this end, we produce the multi-annual China building rooftop area dataset (CBRA) with 2.5 m resolution from 2016-2021 Sentinel-2 images. The CBRA is the first full-coverage and multi-annual BRA data in China. The CBRA achieves good performance with the F1 score of 62.55% (+10.61% compared with the previous BRA data in China) based on 250,000 testing samples in urban areas, and the recall of 78.94% based on 30,000 testing samples in rural areas. 

The CBRA is organized as GeoTIFF (.tif) raster file format with a single band and GCS_WGS_1984 coordinate system. The pixel values are 0 and 255, with 0 representing the background and 255 representing the building rooftop area. Furthermore, to facilitate the use of the data, the CBRA is split into 215 tiles of  spatial grid, named “CBRA_year_E/W**N/S**.tif”, where “year” is the sampling year, the “E/W**N/S**” is the latitude and longitude coordinates found in the upper left corner of the tile data.

The code to generate CBRA can be found here: https://github.com/zpl99/STSR-Seg

高精度且时效性强的建筑屋顶面积（Building Rooftop Area, BRA）地图对于政策制定与可持续发展而言至关重要。此外，作为人类活动的细粒度表征指标，BRA可为城市规划与能源建模提供支撑，进而增进人类福祉。然而，现有大规模BRA数据集（如微软（Microsoft）、谷歌（Google）推出的相关产品）尚未覆盖中国境内，导致国内尚无全覆盖的BRA地图。为此，我们基于2016-2021年Sentinel-2影像，构建了分辨率为2.5米的逐年中国建筑屋顶面积数据集（China Building Rooftop Area Dataset, CBRA）。该数据集是国内首份全覆盖、逐年更新的BRA数据集。基于城市区域25万个测试样本的评估显示，CBRA的F1值达62.55%（较国内此前的BRA数据集提升10.61%）；基于农村区域3万个测试样本的召回率达78.94%，表现优异。

CBRA采用单波段GeoTIFF（.tif）栅格文件格式，坐标系为GCS_WGS_1984。像素值仅包含0与255，其中0代表背景区域，255代表建筑屋顶区域。为便于数据使用，CBRA被划分为215个空间网格瓦片，命名格式为「CBRA_year_E/W*N/S*.tif」，其中「year」为数据采集年份，「E/W*N/S」为该瓦片左上角对应的经纬度坐标。

CBRA的生成代码可通过以下链接获取：https://github.com/zpl99/STSR-Seg