A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER

Mehmet Umut Salur; Şükriye Karaman

doi:10.18038/estubtda.1864067

EN

A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER

Abstract

Early detection of thermal anomalies in photovoltaic (PV) panels is crucial for minimizing energy losses and improving maintenance planning. However, thermal images acquired under real-world field conditions often suffer from low contrast, non-uniform illumination, and noise, which significantly hinder the detection of small-scale faults. In this study, the effects of two image enhancement approaches Zero-Reference Deep Curve Estimation (Zero-DCE) and Swin Transformer for Image Restoration (SwinIR) on a YOLOv11-based object detection model are comparatively analyzed. The Zero-DCE method optimizes illumination and contrast without requiring reference images, whereas the SwinIR model focuses on noise suppression and preservation of structural details. To quantitatively evaluate the impact of both enhancement strategies on YOLOv11 performance, a series of experiments were conducted using identical training and testing protocols. The YOLOv11m model trained on raw thermal images achieved mean mAP@0.5 and mAP@0.5:0.95 scores of 0.849 and 0.748, respectively. When Zero-DCE-enhanced inputs were used, the mAP@0.5 increased to 0.943, while the mean recall reached 0.934, indicating an improvement in anomaly detection sensitivity. In contrast, the SwinIR-enhanced model achieved mean mAP@0.5 and mAP@0.5:0.95 scores of 0.843 and 0.755, respectively, demonstrating a limited improvement at higher IoU thresholds. The experimental results clearly indicate that different image enhancement strategies have a significant influence on fault detection performance. Moreover, the selection of an appropriate enhancement method plays a critical role in reliably identifying small and low-contrast anomalies in thermal PV imagery.

Keywords

Deep Learning, Photovoltaic Fault Detection, YOLOv11, Image Enhancement, Swin Transformer

References

[1] El-Banby GM, Moawad NM, Abouzalm BA, Abouzaid WF, Ramadan EA. Photovoltaic system fault detection techniques: a review. Neural Comput Appl. 2023;35(35):24829-24842. doi:10.1007/s00521-023-09041-7.
[2] IEA. Solar PV. International Energy Agency. Last updated February 3, 2025. Accessed December 25, 2025. https://www.iea.org/energy-system/renewables/solar-pv
[3] Heo S, Lee JH. Fault detection and classification using artificial neural networks. IFAC-PapersOnLine. 2018;51(18):470-475. doi:10.1016/j.ifacol.2018.09.380.
[4] Tas MPM, van Sark WGJHM. Experimental repair technique for glass defects of glass-glass photovoltaic modules - A techno-economic analysis. Solar Energy Materials and Solar Cells. 2023;257:112397. doi:10.1016/j.solmat.2023.112397.
[5] Cao R, Ding Y, Fang X, Liang S, Qi F, Yang Q, Yan W. Photovoltaic array reconfiguration under partial shading conditions based on ant colony optimization. In: 2020 Chinese Control and Decision Conference (CCDC); 22–24 August 2020; Hefei, China. New York, NY, USA: IEEE; 2020:703-708. doi:10.1109/CCDC49329.2020.9164084.
[6] Maghami MR, Hizam H, Gomes C, Radzi MA, Rezadad MI, Hajighorbani S. Power loss due to soiling on solar panel: A review. Renewable and Sustainable Energy Reviews. 2016;59:1307-1316. doi:10.1016/j.rser.2016.01.044.
[7] Sinha A, Sastry OS, Gupta R. Nondestructive characterization of encapsulant discoloration effects in crystalline-silicon PV modules. Solar Energy Materials and Solar Cells. 2016;155:234-242. doi:10.1016/j.solmat.2016.06.019.
[8] Sangpongsanont Y, Chuangchote S, Chenvidhya D, Kirtikara K. Annual expansion in delamination of front encapsulant in tropical climate Field-Operated PV modules. Solar Energy. 2023;262:111850. doi:10.1016/j.solener.2023.111850.
[9] Intermountain Wind & Solar. Does Corrosion Affect PV Solar Performance? Accessed December 25, 2025. https://www.intermtnwindandsolar.com/does-corrosion-affect-pv-solar-performance/
[10] Dhimish M, Tyrrell AM. Power loss and hotspot analysis for photovoltaic modules affected by potential induced degradation. Npj Mater Degrad. 2022;6(1):11. doi:10.1038/s41529-022-00221-9.

[11] Dhimish M, Holmes V, Mehrdadi B, Dales M. The impact of cracks on photovoltaic power performance. Journal of Science: Advanced Materials and Devices. 2017;2(2):199-209. doi:10.1016/j.jsamd.2017.05.005.
[12] Aghaei M, Fairbrother A, Gok A, Ahmad S, Kazim S, Lobato K, Oreski G, Reinders A, Schmitz J, Theelen M, et al. Review of degradation and failure phenomena in photovoltaic modules. Renew Sustain Energy Rev. 2022;159:112160. doi:10.1016/j.rser.2022.112160.
[13] Chang M, Chen C, Hsueh CH, Hsieh WJ, Yen E, Ho KL, Chuang HP, Lee CY, Chen H. The reliability investigation of PV junction box based on 1GW worldwide field database. In: 2015 IEEE 42nd Photovoltaic Specialist Conference (PVSC). IEEE; 2015:1-4. doi:10.1109/PVSC.2015.7356130.
[14] Livera A, Florides M, Theristis M, Makrides G, Georghiou GE. Failure diagnosis of short- and open-circuit fault conditions in PV systems. In: 2018 IEEE 7th World Conference on Photovoltaic Energy Conversion (WCPEC). IEEE; 2018:0739-0744. doi:10.1109/PVSC.2018.8548161.
[15] Taah PT, Asoh DA, Mungwe JN, Ncheuveu NT, Awangum NN. Fault Diagnosis in Photovoltaic Systems Using Machine Learning Algorithms. SSRN preprint. Published online April 21, 2025. doi:10.2139/ssrn.5222406.
[16] Jalal M, Khalil IU, Haq A ul. Deep learning approaches for visual faults diagnosis of photovoltaic systems: State-of-the-Art review. Results in Engineering. 2024;23:102622. doi:10.1016/j.rineng.2024.102622.
[17] Ali ML, Zhang Z. The YOLO Framework: A Comprehensive Review of Evolution, Applications, and Benchmarks in Object Detection. Computers. 2024;13(12):336. doi:10.3390/computers13120336.
[18] Cong X, Li S, Chen F, Liu C, Meng Y. A review of YOLO object detection algorithms based on deep learning. Frontiers in Computing and Intelligent Systems. 2023;4(2):17-20.
[19] Salur MU, Aydin İ, Karaköse M. Real-Time Detection of Threatening Objects with Embedded Deep Learning. DÜMF Mühendislik Dergisi. 2019;10(2):497-509. doi:10.24012/dumf.492433.
[20] Dhimish M. HOTSPOT-YOLO: A Lightweight Deep Learning Attention-Driven Model for Detecting Thermal Anomalies in Drone-Based Solar Photovoltaic Inspections. arXiv preprint. Published online August 26, 2025. doi:10.48550/arXiv.2508.18912.
[21] He H, Duan Y, Liu W, et al. A Self-Evolving Defect Detection Framework for Industrial Photovoltaic Systems. arXiv preprint. Published online March 16, 2026. doi:10.48550/arXiv.2603.14869.
[22] Guo C, Li C, Guo J, Loy CC, Hou J, Kwong S, Cong R. Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE; 2020:1777-1786. doi:10.1109/CVPR42600.2020.00185.
[23] Liu Z, Lin Y, Cao Y, Hu H, Wei Y, Zhang Z, Lin S, Guo B. Swin Transformer: Hierarchical Vision Transformer using Shifted Windows. In: 2021 IEEE/CVF International Conference on Computer Vision (ICCV). IEEE; 2021:9992-10002. doi:10.1109/ICCV48922.2021.00986.
[24] Hassan S, Dhimish M. A Survey of CNN-Based Approaches for Crack Detection in Solar PV Modules: Current Trends and Future Directions. Solar. 2023;3(4):663-683. doi:10.3390/solar3040036.
[25] Deitsch S, Christlein V, Berger S, Buerhop-Lutz C, Maier A, Gallwitz F, Riess C. Automatic classification of defective photovoltaic module cells in electroluminescence images. Sol Energy. 2019;185:455-468. doi:10.1016/j.solener.2019.02.067.
[26] Masita K, Hasan A, Shongwe T, Hilal HA. Deep learning in defects detection of PV modules: A review. Solar Energy Advances. 2025;5:100090. doi:10.1016/j.seja.2025.100090.
[27] Yanılmaz S, Türkoğlu M, Aslan M. Detection of Hotspot Defects in Solar Power Plants with YOLO Algorithms. Fırat Üniversitesi Mühendislik Bilimleri Dergisi. 2024;36(1):121-132. doi:10.35234/fumbd.1318060.
[28] Ab Hamid MZ, Daud K, Che Soh ZH, Osman MK, Isa IS, Jadin MS. Deep Learning-Driven Thermal Imaging Hotspot Detection in Solar Photovoltaic Arrays Using YOLOv10. ESTEEM Academic Journal. 2024;20(September):77-90. doi:10.24191/esteem.v20iSeptember.1861.g1773.
[29] Mohd Mokhtar HS, Abdul Nasir AS, Naim Tajuddin MF, Abdul Nasir MH, Ananda Rao K. Integrating Deep Transfer Learning and Image Enhancement for Enhancing Defective Photovoltaic Cells Classification in Electroluminescence Images. Iranian Journal of Electrical and Electronic Engineering. 2025;21(2):3571.
[30] Joshi D, Pal M. Lightweight Transformer-Driven Segmentation of Hotspots and Snail Trails in Solar PV Thermal Imagery. arXiv preprint. Published online July 28, 2025. arXiv:2507.20680.
[31] Guo Y, Wang X, Lin Z. RFE-YOLO: A Study on Photovoltaic Module Fault Detection Algorithm Based on Multimodal Feature Fusion. Sensors. 2025;25(21):6774. doi:10.3390/s25216774.
[32] Qureshi UR, Rashid A, Altini N, Bevilacqua V, La Scala M. Explainable Intelligent Inspection of Solar Photovoltaic Systems with Deep Transfer Learning: Considering Warmer Weather Effects Using Aerial Radiometric Infrared Thermography. Electronics. 2025;14(4):755. doi:10.3390/electronics14040755.
[33] ZAE Bayern. ELPV Dataset. GitHub. A dataset of functional and defective solar cells extracted from EL images of solar modules. Accessed April 2, 2026. https://github.com/zae-bayern/elpv-dataset.
[34] Darabi PK. Thermal Solar PV Anomaly Detection Dataset. Kaggle. Accessed December 23, 2025. https://www.kaggle.com/datasets/pkdarabi/solarpanel.
[35] Christakakis P, Pechlivani EM, Dimou P. Thermal PV Panel Detection and Fault Detection Dataset for UAV-Based Inspection. Zenodo. 2025. doi:10.5281/zenodo.16420123.
[36] Bello RW, Owolawi PA, van Wyk EA, Du C. Photovoltaic module dataset for automated fault detection and analysis in large photovoltaic systems using photovoltaic module fault detection. Data Brief. 2024;57:111184. doi:10.1016/j.dib.2024.111184.
[37] Abdelshafie Abuhussein MF. Thermal Imaging Dataset for Hotspot Detection on Solar Panels: Impact of Bird Droppings on Efficiency. Mendeley Data. 2024;V1. doi:10.17632/8sxfmrpfpv.1.
[38] Wang B, Chen Q, Wang M, Chen Y, Zhang Z, Liu X, Gao W, Zhang Y, Zhang H. PVF-10: a high-resolution unmanned aerial vehicle thermal infrared image dataset for fine-grained photovoltaic fault classification. Applied Energy. 2024;376:124187. doi:10.1016/j.apenergy.2024.124187.
[39] Khanam R, Hussain M. YOLOv11: An Overview of the Key Architectural Enhancements. arXiv preprint. Published online October 23, 2024. doi:10.48550/arXiv.2410.17725.
[40] Shamisavi M, Segovia Ramirez I, Gómez Muñoz CQ. Comparative Evaluation of YOLO- and Transformer-Based Models for Photovoltaic Fault Detection Using Thermal Imagery. Energies. 2026;19(3):845. doi:10.3390/en19030845.
[41] Afifah ANN, Indrabayu, Suyuti A, Syafaruddin. Improving the image quality of grayscale thermal images taking from photovoltaic panel with contrast enhancement method. International Journal of Innovative Computing, Information and Control. 2023;19(1):197-212. doi:10.24507/ijicic.19.01.197.
[42] Chen RC, Dewi C, Zhuang YC, Chen JK. Contrast Limited Adaptive Histogram Equalization for Recognizing Road Marking at Night Based on Yolo Models. IEEE Access. 2023;11:92926-92942. doi:10.1109/ACCESS.2023.3309410.
[43] Salur MU, Karakas C, Aydin I. Enhancing the Quality of Satellite Images in Disaster Management: A Comparative Analysis of Zero-DCE, CIDNet, and MIRNet Models. Turk Deprem Arastirma Dergisi. 2025;7(1):101-114. doi:10.46464/tdad.1600376.
[44] Ahmad MSZ, Aziz NAA, Lim HS, Ghazali AK, Latiff ‘AA. Impact of Image Enhancement Using Contrast-Limited Adaptive Histogram Equalization (CLAHE), Anisotropic Diffusion, and Histogram Equalization on Spine X-Ray Segmentation with U-Net, Mask R-CNN, and Transfer Learning. Algorithms. 2025;18(12):796. doi:10.3390/a18120796.

Details

Primary Language

English

Subjects

Image Processing, Pattern Recognition, Deep Learning, Knowledge Representation and Reasoning

Journal Section

Research Article

Authors

Mehmet Umut Salur ^*
0000-0003-0296-6266
Türkiye

Şükriye Karaman This is me
0009-0001-8455-6414
Türkiye

Publication Date

June 25, 2026

Submission Date

January 15, 2026

Acceptance Date

May 4, 2026

Published in Issue

Year 2026 Volume: 27 Number: 2

DOI

https://doi.org/10.18038/estubtda.1864067

IZ

https://izlik.org/JA26TJ76BH

APA

Salur, M. U., & Karaman, Ş. (2026). A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER. Eskişehir Technical University Journal of Science and Technology A - Applied Sciences and Engineering, 27(2), 314-337. https://doi.org/10.18038/estubtda.1864067

AMA

1.Salur MU, Karaman Ş. A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER. Estuscience - Se. 2026;27(2):314-337. doi:10.18038/estubtda.1864067

Chicago

Salur, Mehmet Umut, and Şükriye Karaman. 2026. “A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER”. Eskişehir Technical University Journal of Science and Technology A - Applied Sciences and Engineering 27 (2): 314-37. https://doi.org/10.18038/estubtda.1864067.

EndNote

Salur MU, Karaman Ş (June 1, 2026) A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER. Eskişehir Technical University Journal of Science and Technology A - Applied Sciences and Engineering 27 2 314–337.

IEEE

[1]M. U. Salur and Ş. Karaman, “A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER”, Estuscience - Se, vol. 27, no. 2, pp. 314–337, June 2026, doi: 10.18038/estubtda.1864067.

ISNAD

Salur, Mehmet Umut - Karaman, Şükriye. “A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER”. Eskişehir Technical University Journal of Science and Technology A - Applied Sciences and Engineering 27/2 (June 1, 2026): 314-337. https://doi.org/10.18038/estubtda.1864067.

JAMA

1.Salur MU, Karaman Ş. A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER. Estuscience - Se. 2026;27:314–337.

MLA

Salur, Mehmet Umut, and Şükriye Karaman. “A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER”. Eskişehir Technical University Journal of Science and Technology A - Applied Sciences and Engineering, vol. 27, no. 2, June 2026, pp. 314-37, doi:10.18038/estubtda.1864067.

Vancouver

1.Mehmet Umut Salur, Şükriye Karaman. A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER. Estuscience - Se. 2026 Jun. 1;27(2):314-37. doi:10.18038/estubtda.1864067

A YOLOv11-BASED APPROACH FOR THERMAL FAULT DETECTION IN PHOTOVOLTAIC PANELS USING ZERO-DCE AND SWIN TRANSFORMER

Abstract

Keywords

Abstract

References

Details

Primary Language

Subjects

Journal Section

Authors

Publication Date

Submission Date

Acceptance Date

Published in Issue

DOI

IZ

Cite