CSPPNet: A Convolution and State-Space-Based Photovoltaic Panel Extraction Network Using Gaofen-2 High-Resolution Imagery

With the rapid development of photovoltaic (PV) industry, it is crucial to accurately identify PV panels using remote sensing data. However, the existing methods still face problems, such as difficulty in distinguishing PV panels from easily confused ground objects, such as dark buildings, roads, an...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenqing Liu, Hongtao Huo, Luyan Ji, Yongchao Zhao, Xiaowen Liu, Jialei Xie
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10892037/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849393882150731776
author Wenqing Liu
Hongtao Huo
Luyan Ji
Yongchao Zhao
Xiaowen Liu
Jialei Xie
author_facet Wenqing Liu
Hongtao Huo
Luyan Ji
Yongchao Zhao
Xiaowen Liu
Jialei Xie
author_sort Wenqing Liu
collection DOAJ
description With the rapid development of photovoltaic (PV) industry, it is crucial to accurately identify PV panels using remote sensing data. However, the existing methods still face problems, such as difficulty in distinguishing PV panels from easily confused ground objects, such as dark buildings, roads, and black plastic, lacking the analysis of the directional characteristics, and inadequate capturing the global dependencies. To address these challenges, a convolution and state-space-based photovoltaic panel extraction network (CSPPNet) is proposed. Specifically, we construct a PV panel index based on visible and near-infrared bands to improve the separability of the PV panels from those easily confused ground objects. Second, considering the unique horizontal characteristics exhibited by the PV panels in remote sensing images, a south-facing orientation prior module is designed to enhance the horizontal features and improve our network in capturing horizontal objects. Finally, the encoder of our network adopts a parallel structure of depthwise separable convolution and state-space module to capture local detailed features and global semantic features of PV panels layer by layer. Furthermore, we build a high-quality dataset [named Photovoltaic Panels in the Eastern and Western Regions of China (PPEWRC)] containing four bands ranging from visible to near-infrared wavelength based on Gaofen-2 satellite images. The experimental results show that our proposed CSPPNet effectively reduces the misjudgment of PV panels, with high completeness and clear edges. The intersection over union, precision, recall, and <italic>F</italic>1-score can reach 77.37&#x0025;, 87.42&#x0025;, 85.11&#x0025;, and 86.98&#x0025;, respectively, on the PPEWRC dataset. Ablation experiments validate the effectiveness of each module and fusion process, providing insights into the sustainable utilization of renewable energy.
format Article
id doaj-art-8fdef4cceb2d457bb861ef0750e54bbb
institution Kabale University
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-8fdef4cceb2d457bb861ef0750e54bbb2025-08-20T03:40:15ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01187644766110.1109/JSTARS.2025.354349010892037CSPPNet: A Convolution and State-Space-Based Photovoltaic Panel Extraction Network Using Gaofen-2 High-Resolution ImageryWenqing Liu0Hongtao Huo1https://orcid.org/0000-0002-1552-4400Luyan Ji2https://orcid.org/0000-0001-5369-4200Yongchao Zhao3Xiaowen Liu4Jialei Xie5https://orcid.org/0009-0007-0223-0554Department of Information and Cyber Security, People's Public Security University of China, Beijing, ChinaDepartment of Information and Cyber Security, People's Public Security University of China, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaDepartment of Information and Cyber Security, People's Public Security University of China, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaWith the rapid development of photovoltaic (PV) industry, it is crucial to accurately identify PV panels using remote sensing data. However, the existing methods still face problems, such as difficulty in distinguishing PV panels from easily confused ground objects, such as dark buildings, roads, and black plastic, lacking the analysis of the directional characteristics, and inadequate capturing the global dependencies. To address these challenges, a convolution and state-space-based photovoltaic panel extraction network (CSPPNet) is proposed. Specifically, we construct a PV panel index based on visible and near-infrared bands to improve the separability of the PV panels from those easily confused ground objects. Second, considering the unique horizontal characteristics exhibited by the PV panels in remote sensing images, a south-facing orientation prior module is designed to enhance the horizontal features and improve our network in capturing horizontal objects. Finally, the encoder of our network adopts a parallel structure of depthwise separable convolution and state-space module to capture local detailed features and global semantic features of PV panels layer by layer. Furthermore, we build a high-quality dataset [named Photovoltaic Panels in the Eastern and Western Regions of China (PPEWRC)] containing four bands ranging from visible to near-infrared wavelength based on Gaofen-2 satellite images. The experimental results show that our proposed CSPPNet effectively reduces the misjudgment of PV panels, with high completeness and clear edges. The intersection over union, precision, recall, and <italic>F</italic>1-score can reach 77.37&#x0025;, 87.42&#x0025;, 85.11&#x0025;, and 86.98&#x0025;, respectively, on the PPEWRC dataset. Ablation experiments validate the effectiveness of each module and fusion process, providing insights into the sustainable utilization of renewable energy.https://ieeexplore.ieee.org/document/10892037/Horizontal featuresphotovoltaic (PV) panel extractionremote sensing imagesstate space
spellingShingle Wenqing Liu
Hongtao Huo
Luyan Ji
Yongchao Zhao
Xiaowen Liu
Jialei Xie
CSPPNet: A Convolution and State-Space-Based Photovoltaic Panel Extraction Network Using Gaofen-2 High-Resolution Imagery
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Horizontal features
photovoltaic (PV) panel extraction
remote sensing images
state space
title CSPPNet: A Convolution and State-Space-Based Photovoltaic Panel Extraction Network Using Gaofen-2 High-Resolution Imagery
title_full CSPPNet: A Convolution and State-Space-Based Photovoltaic Panel Extraction Network Using Gaofen-2 High-Resolution Imagery
title_fullStr CSPPNet: A Convolution and State-Space-Based Photovoltaic Panel Extraction Network Using Gaofen-2 High-Resolution Imagery
title_full_unstemmed CSPPNet: A Convolution and State-Space-Based Photovoltaic Panel Extraction Network Using Gaofen-2 High-Resolution Imagery
title_short CSPPNet: A Convolution and State-Space-Based Photovoltaic Panel Extraction Network Using Gaofen-2 High-Resolution Imagery
title_sort csppnet a convolution and state space based photovoltaic panel extraction network using gaofen 2 high resolution imagery
topic Horizontal features
photovoltaic (PV) panel extraction
remote sensing images
state space
url https://ieeexplore.ieee.org/document/10892037/
work_keys_str_mv AT wenqingliu csppnetaconvolutionandstatespacebasedphotovoltaicpanelextractionnetworkusinggaofen2highresolutionimagery
AT hongtaohuo csppnetaconvolutionandstatespacebasedphotovoltaicpanelextractionnetworkusinggaofen2highresolutionimagery
AT luyanji csppnetaconvolutionandstatespacebasedphotovoltaicpanelextractionnetworkusinggaofen2highresolutionimagery
AT yongchaozhao csppnetaconvolutionandstatespacebasedphotovoltaicpanelextractionnetworkusinggaofen2highresolutionimagery
AT xiaowenliu csppnetaconvolutionandstatespacebasedphotovoltaicpanelextractionnetworkusinggaofen2highresolutionimagery
AT jialeixie csppnetaconvolutionandstatespacebasedphotovoltaicpanelextractionnetworkusinggaofen2highresolutionimagery