Abstract
Accurate segmentation of photovoltaic (PV) panels from remote sensing imagery is essential for power plant management, yet remains challenging due to complex backgrounds and high reflectivity. To address this, we propose DBGENet, a unified network integrating global context modeling, local detail preservation, and edge-aware learning. Specifically, a dual-branch encoder collaboratively extracts global semantics and fine-grained spatial features. This architecture is enhanced by a global channel-spatial attention module to suppress background interference, and an edge-aware lightweight bottleneck module to incorporate explicit boundary priors for improved contour integrity. Extensive experiments on the publicly available Earth System Science Data (ESSD) benchmark, which comprises UAV (PV01, 0.1 m), aerial (PV03, 0.3 m), and satellite (PV08, 0.8 m) imagery, demonstrate DBGENet’s superiority. Our method achieves an mIoU of 94.48% for rooftop PV01, 96.63% for ground-mounted PV03 scenarios such as water surfaces, 80.13% for satellite PV08, exhibiting exceptional boundary accuracy and robustness across diverse complex scenes.
Funding Information
  • The Opening Fund of Intelligent Perception and Control Key Laboratory of Sichuan Province (2025IPCY02)
  • the Innovation Fund of Postgraduate, Sichuan University of Science and Engineering (Y2025090)
  • the Sichuan province natural science fund project (2025ZNSFSC0477)
  • the Open Research Fund of Sichuan Province Key Laboratory of Intelligent Medicine and Health Big Data (ZNYX2507)

This publication has 31 references indexed in Scilit: