BHDSI：面向深度学习的遥感建筑高度数据集（5606）

王浩; 马遥; 曹昌昊; 宁晓刚; 张翰超; 张瑞倩

doi:10.11834/jrs.20255103

浏览量 : 0 下载量: 1055 CSCD: 0

R-PDF
PDF
导出
分享
收藏
专辑

BHDSI：面向深度学习的遥感建筑高度数据集（5606）
BHDSI： A Remote Sensing Building Height Dataset for Deep Learning
2025年页码：1-13
收稿：2025-04-02，

网络首发：2025-10-28，
DOI： 10.11834/jrs.20255103
稿件说明：

移动端阅览

王浩，马遥，曹昌昊，宁晓刚，张翰超，张瑞倩.XXXX.BHDSI：面向深度学习的遥感建筑高度数据集（5606）.遥感学报，XX（XX）： 1-13 DOI： 10.11834/jrs.20255103.

WANG Hao，MA Yao，CAO Changhao，NING Xiaogang，ZHANG Hanchao，ZHANG Ruiqian. XXXX. BHDSI： A Remote Sensing Building Height Dataset for Deep Learning. National Remote Sensing Bulletin， XX（XX）：1-13 DOI： 10.11834/jrs.20255103.

摘要

针对当前中国区域建筑高度遥感数据匮乏的问题，本文构建了一个面向深度学习建筑高度估计任务的大规模数据集。利用光学和SAR遥感影像进行建筑高度估计对于理解城市形态和优化城市存量空间具有重要意义。然而，现有的数据集存在诸多局限：由于样本数量较少，难以满足基于深度学习的遥感信息提取需求，样本所覆盖的区域较为有限，无法提供足够的地理多样性和空间特征代表性，特别是针对中国区域的大规模建筑高度数据集尤为缺乏。此外，数据集的开源性不足，限制了其在更广泛的研究中的应用和验证。为解决这些问题，本文构建了一个专注于建筑高度回归的数据集BHDSI，涵盖了中国62个城市的中心城区，共有5606个样本，覆盖了城市，农村等场景，是目前中国区域覆盖面积最大的建筑高度数据集。该数据集包含哨兵-1和哨兵-2的遥感影像以及建筑高度的真实值，样本大小是256×256，相比于64×64大小的数据集，为建筑高度估计研究提供了一个重要的补充选择。相比其他数据集，该数据集具有样本数量大、覆盖范围广、建筑高度分布合理等特点，能够更好地满足深度学习网络的训练需求。在此基础上，本文采用相同的深度学习网络对BHDSI数据集及其他类似数据集进行了评估，并对比了多个网络使用BHDSI数据集时在建筑高度回归任务中的表现，深入分析了各网络的优劣。研究表明，与其他数据集相比，BHDSI数据集在建筑高度回归任务中的表现更加优异。进一步分析发现，使用BHDSI数据集时，建筑高度较低的区域其估计精度相对较高。此外，U-Net解码器用于建筑高度估计网络训练能够取得更高的精度。这一数据集及实验结果为未来建筑高度估计领域的研究提供了重要的支持。

Abstract

Objective The objective of this study is to address the limitations of existing datasets used for building height estimation from optical and SAR remote sensing imagery. Current datasets often suffer from small sample sizes

limited geographic diversity

and a lack of openness

making them insufficient for supporting deep learning-based remote sensing applications—especially for large-scale studies in China. Accurately estimating building heights is critical for understanding urban morphology and optimizing urban stock space

thus necessitating the development of a more comprehensive

representative

and accessible dataset. Methods To overcome these issues

this paper constructs a new dataset named BHDSI

specifically designed for building height regression tasks. The dataset comprises 5

606 samples from the central urban areas of 62 cities across China

making it the largest building height dataset for the country in terms of geographic coverage. It includes both Sentinel-1 and Sentinel-2 imagery along with true building height values

with each sample having a spatial resolution of 256×256 pixels. This provides a richer spatial context compared to existing datasets with smaller sample sizes

such as 64×64. The dataset encompasses a wide range of scenarios

including urban and rural areas

ensuring better representation of spatial features. Result Experimental evaluations demonstrate that the BHDSI dataset leads to superior performance in building height regression tasks when compared to other similar datasets

across various deep learning networks. The results also indicate that estimation accuracy tends to be higher in regions with lower building heights. Furthermore

the study finds that using a U-Net decoder structure in the network architecture contributes to higher prediction precision

highlighting the importance of decoder design in deep learning-based height estimation. Conclusion The BHDSI dataset significantly advances the field of building height estimation by offering a large-scale

diverse

and high-quality resource tailored for deep learning. Its broad coverage

balanced height distribution

and open accessibility make it better suited for training and evaluating deep neural networks than previously available datasets. The study confirms that both data quality and network architecture

especially decoder design

play vital roles in improving estimation accuracy

and BHDSI serves as a strong foundation for future research in this domain.

关键词

Keywords

references

Cai B. , Shao Z. , Huang X. , Zhou X. , and Fang S . 2023 . Deep learning-based building height mapping using Sentinel-1 and Sentinel-2 data . International Journal of Applied Earth Observation and Geoinformation , 122 . [ DOI: 10.1016/j.jag.2023.103399 http://dx.doi.org/10.1016/j.jag.2023.103399 ]

Cao Y. , and Huang X . ( 2021 ). A deep learning method for building height estimation using high-resolution multi-view imagery over urban areas: A case study of 42 Chinese cities . Remote Sensing of Environment , 264 , 112590 . [ DOI: 10.1016/j.rse.2021.112590 http://dx.doi.org/10.1016/j.rse.2021.112590 ]

Cao Y. , and Weng Q . 2024 . A deep learning-based super-resolution method for building height estimation at 2.5m spatial resolution in the Northern Hemisphere . Remote Sensing of Environment , 310 . [ DOI: doi.org/10.1016/j.rse.2024.114241 http://dx.doi.org/doi.org/10.1016/j.rse.2024.114241 ]

Chen L. C. , Papandreou G. , Schroff F. , and Adam H . 2017 . Rethinking Atrous Convolution for Semantic Image Segmentation . [ DOI: 10.48550/arXiv.1706.05587 http://dx.doi.org/10.48550/arXiv.1706.05587 ]

He K. , Zhang X. , Ren S. , and Sun J . 2016 . Deep Residual Learning for Image Recognition . IEEE . [ DOI: 10.1109/CVPR.2016.90 http://dx.doi.org/10.1109/CVPR.2016.90 ]

Koppel K. , Zalite K. , Voormansik K. , and Jagdhuber T . 2017 . Sensitivity of Sentinel-1 backscatter to characteristics of buildings . International Journal of Remote Sensing , 38 ( 22 ), 6298 - 6318 . [ DOI: 10.1080/01431161.2017.1353160 http://dx.doi.org/10.1080/01431161.2017.1353160 ]

Kumar L. , and Mutanga O . 2018 . Google Earth Engine Applications Since Inception: Usage, Trends, and Potential . Remote Sensing of Environment , 10 ( 10 ). [ DOI: 10.3390/rs10101509 http://dx.doi.org/10.3390/rs10101509 ]

Liu Z. , Tang H. , Feng L. , and Lyu S . 2023 . China Building Rooftop Area: 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 . Earth System Science Data , 15 ( 8 ), 3547 - 3572 . [ DOI: 10.5194/essd-15-3547-2023 http://dx.doi.org/10.5194/essd-15-3547-2023 ]

Luo W. , Li Y. , Urtasun R. , and Zemel R . 2017 . Understanding the Effective Receptive Field in Deep Convolutional Neural Networks .

Persello C. , Hänsch R. , Vivone G. , Chen K. , Yan Z. , Tang D. , . . . Sun , X . 2023 . 2023 IEEE GRSS Data Fusion Contest: Large-Scale Fine-Grained Building Classification for Semantic Urban Reconstruction [Technical Committees ] . IEEE Geoscience and Remote Sensing Magazine, 11 ( 1 ), 94 - 97 . [ DOI: 10.1109/mgrs.2023.3240233 http://dx.doi.org/10.1109/mgrs.2023.3240233 ]

Richter M. L. , Byttner W. , Krumnack U. , Wiedenroth A. , Schallner L. , and Shenk J . 2021 . (Input) Size Matters for CNN Classifiers . ArXiv . [ DOI: 10.1007/978-3-030-86340-1_11 http://dx.doi.org/10.1007/978-3-030-86340-1_11 ]

Robinson J. , Chuang C . -Y. , Sra S. , and Jegelka S. J . A. 2020 . Contrastive Learning with Hard Negative Samples. abs/2010. 04592 . [ DOI: 10.48550/arXiv.2010.04592 http://dx.doi.org/10.48550/arXiv.2010.04592 ]

Ronneberger O. , Fischer P. , and Brox T . 2015 . U-Net: Convolutional Networks for Biomedical Image Segmentation . International Conference on Medical Image Computing and Computer-Assisted Intervention ,

Simonyan K. , and Zisserman A . 2014 . Very Deep Convolutional Networks for Large-Scale Image Recognition . Computer Science . [ DOI: 10.48550/arXiv.1409.1556 http://dx.doi.org/10.48550/arXiv.1409.1556 ]

Stephane , Lathuiliere , Pablo , Mesejo , Xavier , Alameda-Pineda , . . . Intelligence , M . 2019 . A Comprehensive Analysis of Deep Regression. [ DOI: 10.1109/TPAMI.2019.2910523 http://dx.doi.org/10.1109/TPAMI.2019.2910523 ]

Tan M. , and Le Q . V . 2019 . EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. [ DOI: 10.48550/arXiv.1905.11946 http://dx.doi.org/10.48550/arXiv.1905.11946 ]

Michail D. , Davalas C. , Panagiotou L . -I. , Prapas I. , Kondylatos S., Bountos N. I., and Papoutsis I. 2025 . FireCastNet: Earth-as-a-Graph for Seasonal Fire Prediction. Computer Science. [ DOI: 10.48550/arXiv.2502.01550 http://dx.doi.org/10.48550/arXiv.2502.01550 ]

Wojna Z. , VittorioGuadarrama , SergioSilberman , NathanChen , Liang-ChiehFathi , AlirezaUijlings , Jasper %J International Journal of Computer Vision . 2019 . The Devil is in the Decoder: Classification, Regression and GANs . International Journal of Computer Vision , 127(11a12 ) . [ DOI: 10.48550/arXiv.1707.05847 http://dx.doi.org/10.48550/arXiv.1707.05847 ]

Xiao T. , Liu Y. , Zhou B. , Jiang Y. , and Sun J . 2018 . Unified Perceptual Parsing for Scene Understanding . European Conference on Computer Vision . [ DOI: 10.48550/arXiv.1807.10221 http://dx.doi.org/10.48550/arXiv.1807.10221 ]

Yinxia C. , and Xin H . 2023 . A full-level fused cross-task transfer learning method for building change detection using noise-robust pretrained networks on crowdsourced labels . Remote Sensing of Environment , 284 . [ DOI: 10.1016/J.RSE.2022.113371 http://dx.doi.org/10.1016/J.RSE.2022.113371

Zhang L. , Xia G . -S., Wu T., Lin L., and Tai X . C. 2016 . Deep Learning for Remote Sensing Image Understanding. Journal of Sensors , 2016, 1 - 2 . [ DOI: 10.1155/2016/7954154 http://dx.doi.org/10.1155/2016/7954154 ]

Zhou C. , Wang Z. , Chen Q. , Jiang Y. , and Pei J . 2014 . Design optimization and field demonstration of natural ventilation for high-rise residential buildings . Energy Buildings , 82 , 457 - 465 . [ DOI: 10.1016/j.enbuild.2014.06.036 http://dx.doi.org/10.1016/j.enbuild.2014.06.036 ]

Wu W , Ma J , Banzhaf E , Meadows M E. , Yu Z , Guo F , Sengupta D , Cai X & Zhao B . 2023 . A first Chinese building height estimate at 10 m resolution (CNBH-10 m) using multi-source earth observations and machine learning . Remote Sensing of Environment .[ DOI: 10.1016/j.rse.2023.113578 http://dx.doi.org/10.1016/j.rse.2023.113578 ]

Feng Q L , Chen B A , Li G Q , Yao X C , Gao B B , Zhang L C . 2022 . A Review of Research on Remote Sensing Image Sample Datasets . Journal of Remote Sensing , 26 ( 4 ), 17 .

冯权泷 , 陈泊安 , 李国庆 , 姚晓闯 , 高秉博 , 张连翀 . 2022 . 遥感影像样本数据集研究综述 . 遥感学报 , 26 ( 4 ), 17 .[ DOI: 10.11834/jrs.20221162 http://dx.doi.org/10.11834/jrs.20221162 ]

Hu K , Wu G Q , Hu Z H , and Wang Z M . 2024 . Research on Metal Surface Defect Image Classification Based on Improved VGG16 Network . Computer Applications and Software .

胡坤 , 吴国庆 , 胡祖辉 , 王忠明 . 2024 . 基于改进的 VGG16 网络金属表面缺陷图像分类研究 . 计算机应用与软件 .[ DOI: 10．3969/j．issn．1000-386x．2024．06．026 http://dx.doi.org/10．3969/j．issn．1000-386x．2024．06．026 ]

Zhang S . 2023 . Urban Building Area Extraction by Fusing Sentinel-1 and Sentinel-2 Remote Sensing Data. 2022/2023 China Urban Planning Annual Conference, Wuhan, Hubei, China .

张澍 . 2023 . 融合Sentinel-1和Sentinel-2遥感数据的城市建筑面积提取 . 2022 / 2023 中国城市规划年会, 中国湖北武汉.

Zhang Y , Meng D C , Song L W , and Dong H L . 2024 . Seismic Velocity Inversion Method Based on Feature-Enhanced U-Net . Petroleum Geophysical Exploration .

张岩 , 孟德聪 , 宋利伟 , 董宏丽 . 2024 . 基于特征强化U-Net的地震速度反演方法 . 石油地球物理勘探 . [ DOI: 10.13810/j.cnki.issn.1000-7210.2024.02.001 http://dx.doi.org/10.13810/j.cnki.issn.1000-7210.2024.02.001 ]

Sun Xun , Zhu Meining , Song Jinling , Liu YongandZhang Sixuan 2024 . Remote sensing image water body recognition based on CBAM and Unet Environmental Science Guide ( 05 ), 91 - 96 .

孙逊 , 祝美宁 , 宋金玲 , 刘勇，张思萱 . 2024 . 基于CBAM和Unet的遥感影像水体识别 . 环境科学导刊 ( 05 ), 91 - 96 .[ DOI: 10.13623/j.cnki.hkdk.2024.05.020 http://dx.doi.org/10.13623/j.cnki.hkdk.2024.05.020 ]

Bai Qian , Luo Xiaobo , and Mu Shilin High resolution remote sensing image surface water extraction based on TA-UNet3+ Computer Engineering and Applications , 1 - 13

白倩，罗小波，母仕林 . 基于TA-UNet3+的高分辨率遥感图像地表水体提取 . 计算机工程与应用 , 1 - 13 .

Zhao Xingwang , Wu Zhiguo , Liu Chao , Liu ChunyangandChen Jian . 2024 Research on Building Extraction Based on CBAM VGG16 UNet Semantic Segmentation Model Journal of Qiqihar University (Natural Science Edition) ( 03 ), 34 - 40 .

赵兴旺 , 吴治国 , 刘超 , 刘春阳，陈健 . 2024 . 基于CBAM VGG16-UNet语义分割模型的建筑物提取研究 . 齐齐哈尔大学学报(自然科学版) ( 03 ), 34 - 40 . [ DOI: 10.20171/j.cnki.23-1419/n.2024.03.004 http://dx.doi.org/10.20171/j.cnki.23-1419/n.2024.03.004 ]