This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessArticle
DSFA-SwinNet: A Multi-Scale Attention Fusion Network for Photovoltaic Areas Detection
by
Shaofu Lin
Shaofu Lin
Prof. Dr. Shaofu Lin is a Professor at the College of Computer Science, Beijing University of China. [...]
Prof. Dr. Shaofu Lin is a Professor at the College of Computer Science, Beijing University of Technology, China. He holds a Ph.D. from the Institute of Remote Sensing and Geographic Information Systems, Peking University. His research areas are 1. smart city spatial-temporal big data; 2. big data computing and intelligence; 3. blockchain and data governance; and 4. internet of things (IoT) and IntelliSense. He is also a distinguished member of China Computer Federation; a senior member of the Chinese Institute of Electronics; an executive member of China Computer Federation Data Governance Development Committee; an executive member of the Block-Chain Commission, and senior member of China Computer Federation; a member of Federated Data and Federated Intelligence Professional Committee, Chinese Association of Automation; a member of BRICS Remote Sensing Satellite Constellation Expert Group; a member of Metaverse Committee of 100, China Science and Technology Communications Center and Chinese Society for Science and Technology Journalism; an executive director of the Disability Statistics Branch of National Statistical Society of China; an executive member of Rehabilitation Engineering and Assistive Technology Professional Committee, China Association of Rehabilitation of Disabled Persons; a vice president of Beijing Information and Telecommunication Association; and a board member of Beijing Institute of Bigdata.
,
Yang Yang
Yang Yang
Xiliang Liu
Xiliang Liu
Li Tian
Li Tian
Dr. Li Tian currently works as an Associate Researcher and Master's Supervisor at the Institute of a [...]
Dr. Li Tian currently works as an Associate Researcher and Master's Supervisor at the Institute of Geographic Sciences and Natural Resources Research, University of Chinese Academy of Sciences. She obtained a Ph.D. from the School of Life Sciences, Sun Yat-sen University in 2013. She completed a postdoctoral fellowship at the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences in 2015. From 2015 to 2017, she was a visiting scholar at the School of Geography, University of Maryland, USA. Her research fields and directions are landscape ecology and integrated physical geography.
1
College of Computer Science, Beijing University of Technology, Chaoyang District, Beijing 100124, China
2
The Institute of Geographic Sciences and Natural Resources Research, University of Chinese Academy of Sciences, Beijing 100101, China
*
Author to whom correspondence should be addressed.
Remote Sens. 2025, 17(2), 332; https://doi.org/10.3390/rs17020332 (registering DOI)
Submission received: 7 November 2024
/
Revised: 6 January 2025
/
Accepted: 17 January 2025
/
Published: 18 January 2025
Abstract
Precise statistics on the spatial distribution of photovoltaics (PV) are essential for advancing the PV industry, and integrating remote sensing with artificial intelligence technologies offers a robust solution for accurate identification. Currently, numerous studies focus on the detection of single-type PV installations through aerial or satellite imagery. However, due to the variability in scale and shape of PV installations in complex environments, the detection results often fail to capture detailed information and struggle to scale for multi-scale PV systems. To tackle these challenges, a detection method known as Dynamic Spatial-Frequency Attention SwinNet (DSFA-SwinNet) for multi-scale PV areas is proposed. First, this study proposes the Dynamic Spatial-Frequency Attention (DSFA) mechanism, the Pyramid Attention Refinement (PAR) bottleneck structure, and optimizes the feature propagation method to achieve dynamic decoupling of the spatial and frequency domains in multi-scale representation learning. Secondly, a hybrid loss function has been developed with weights optimized employing the Bayesian Optimization algorithm to provide a strategic method for parameter tuning in similar research. Lastly, the fixed window size of Swin-Transformer is dynamically adjusted to enhance computational efficiency and maintain accuracy. The results on two PV datasets demonstrate that DSFA-SwinNet significantly enhances detection accuracy and scalability for multi-scale PV areas.
Share and Cite
MDPI and ACS Style
Lin, S.; Yang, Y.; Liu, X.; Tian, L.
DSFA-SwinNet: A Multi-Scale Attention Fusion Network for Photovoltaic Areas Detection. Remote Sens. 2025, 17, 332.
https://doi.org/10.3390/rs17020332
AMA Style
Lin S, Yang Y, Liu X, Tian L.
DSFA-SwinNet: A Multi-Scale Attention Fusion Network for Photovoltaic Areas Detection. Remote Sensing. 2025; 17(2):332.
https://doi.org/10.3390/rs17020332
Chicago/Turabian Style
Lin, Shaofu, Yang Yang, Xiliang Liu, and Li Tian.
2025. "DSFA-SwinNet: A Multi-Scale Attention Fusion Network for Photovoltaic Areas Detection" Remote Sensing 17, no. 2: 332.
https://doi.org/10.3390/rs17020332
APA Style
Lin, S., Yang, Y., Liu, X., & Tian, L.
(2025). DSFA-SwinNet: A Multi-Scale Attention Fusion Network for Photovoltaic Areas Detection. Remote Sensing, 17(2), 332.
https://doi.org/10.3390/rs17020332
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article Access Statistics
For more information on the journal statistics, click
here.
Multiple requests from the same IP address are counted as one view.
