EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TP-YOLO: Thermal-Prior-Enhanced Photovoltaic Anomaly Detection in UAV Thermal Infrared Imagery

Feng Xing, Yuchuan Yang, Deng Pan, Caiyan Qin

Remote Sensing · 04.11.2026 · 18 (21) · 3789

DOI: 10.3390/rs18213789 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 04.11.2026. Verfasst von Feng Xing und 3 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shenzhen Municipal Science and Technology Innovation Bureau, National Natural Science Foundation of China, Jinzhou 2025 Guided Science and Technology Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unmanned aerial vehicle (UAV) thermal infrared remote sensing can rapidly acquire thermal-response information for large-scale photovoltaic (PV) arrays. Nevertheless, local thermal anomalies occupy small regions within aerial images; their features tend to degrade during successive down-sampling. Moreover, model deployment on edge devices is constrained by limited computational and memory resources. To address these challenges, this paper proposes TP-YOLO, a photovoltaic anomaly-detection approach, and implements it on the Zynq-7020 hardware platform. First, a three-channel thermal-prior input consisting of luminance distribution, local luminance difference, and a luminance gradient is constructed from pseudo-color infrared images, and explicitly characterizes thermal anomalies without requiring absolute-temperature measurements. Next, the detection scales of YOLOv11n are adjusted to P2/P3/P4. Modules including A-C3k2, Conv + Add, and TLCB are adopted to optimize feature extraction and multi-scale fusion, preserving local anomaly features while restraining model complexity. On the hardware side, a PS/PL co-processing architecture is employed. A dedicated convolution IP, ADD IP, and Window IP are designed for convolution, element-wise fusion, and local-window operations, respectively. INT8 quantization and ping-pong buffering are further utilized to realize network deployment. On the PV HSD test set, TP-YOLO achieves an mAP@0.5 of 77.5% and an mAP@0.5:0.95 of 33.7%, with 1.91 M parameters and 6.2 GFLOPs computational cost. In cross-dataset evaluation on the Drone Dataset without fine-tuning, the mAP@0.5 reaches 64.4%. After deploying the INT8 quantized model onto Zynq-7020, the implemented system yields an mAP@0.5 of 76.8%, achieving single-frame inference latency of 162 ms and throughput of 6.17 FPS at 150 MHz, with a power consumption of 3.4 W.

Förderung

  • Shenzhen Municipal Science and Technology Innovation Bureau
  • National Natural Science Foundation of China
  • Jinzhou 2025 Guided Science and Technology Program

Zitieren

APA

Xing, F., Yang, Y., Pan, D., & Qin, C. (2026). TP-YOLO: Thermal-Prior-Enhanced Photovoltaic Anomaly Detection in UAV Thermal Infrared Imagery. Remote Sensing, 18(21), 3789. https://doi.org/10.3390/rs18213789

Vancouver

Xing F, Yang Y, Pan D, Qin C. TP-YOLO: Thermal-Prior-Enhanced Photovoltaic Anomaly Detection in UAV Thermal Infrared Imagery. Remote Sensing. 2026;18(21):3789. doi:10.3390/rs18213789

BibTeX
@article{Xing2026,
  author = {Xing, Feng and Yang, Yuchuan and Pan, Deng and Qin, Caiyan},
  title = {TP-YOLO: Thermal-Prior-Enhanced Photovoltaic Anomaly Detection in UAV Thermal Infrared Imagery},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {21},
  pages = {3789},
  doi = {10.3390/rs18213789}
}
Kostenlos starten →