Araştırma Makalesi

Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset

Cilt: 13 Sayı: 1 21 Ekim 2026
Fatih Gündoğdu , Mücahit Çalışan *
PDF İndir
Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi Kapak Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi
EN TR

Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset

Öz

Cell defects in photovoltaic (PV) modules directly affect production yield, making early and reliable detection critical. Electroluminescence (EL) imaging is widely used for this purpose since it reveals defects such as microcracks and inactive regions; however, manual assessment is time-consuming and subjective. In this study, EL images from the public ELPV benchmark dataset are automatically classified into four defect-severity levels (flawless, low, medium, high). The proposed method augments a ResNet50 backbone with the Squeeze-and-Excitation (SE) block, which adaptively recalibrates channel-wise features. The model is compared against ResNet50, MobileNetV2 and EfficientNetB0 baselines using ImageNet pretrained weights and stratified 5-fold cross-validation. The proposed SE-ResNet50 outperforms all baselines with 0.7609 accuracy and 0.559 macro-F1. Furthermore, an ordinal error-magnitude analysis shows that the SE block yields the smallest and closest errors, reducing the mean absolute level error to 0.443. Nevertheless, the contribution of SE remains limited for the minority 'medium' class, indicating that class imbalance is the principal bottleneck.

Anahtar Kelimeler

Electroluminescence, Photovoltaic, Squeeze-and-Excitation, ResNet50

Kaynakça

  1. Doğan F, Aktaş M, Olgun N, Çalışan M, Türkoğlu İ. Detection of Cell Damage and Types in Solar Panels: A Comprehensive Review. Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi 2026;17(1). https://doi.org/10.24012/dumf.1704433
  2. Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Communications of the ACM 2017;60(6):84–90. https://doi.org/10.1145/3065386
  3. Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition. In: 3rd International Conference on Learning Representations (ICLR); 2015; San Diego, USA. https://doi.org/10.48550/arXiv.1409.1556
  4. 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. Solar Energy 2019;185:455-468. https://doi.org/10.1016/j.solener.2019.02.067
  5. Tang W, Yang Q, Xiong K, Yan W. Deep learning based automatic defect identification of photovoltaic module using electroluminescence images. Solar Energy 2020;201:453-460. https://doi.org/10.1016/j.solener.2020.03.049
  6. Akram MW, Li G, Jin Y, Chen X, Zhu C, Zhao X, Khaleel A. CNN based automatic detection of photovoltaic cell defects in electroluminescence images. Energy 2019;189:116319. https://doi.org/10.1016/j.energy.2019.116319
  7. Buerhop-Lutz C, Deitsch S, Maier A, Gallwitz F, Berger S, Christlein V, Riess C. A benchmark for visual identification of defective solar cells in electroluminescence imagery. In: 35th European Photovoltaic Solar Energy Conference and Exhibition (EU PVSEC); 2018; Brussels, Belgium. https://doi.org/10.4229/35thEUPVSEC20182018-5CV.3.15
  8. Hu J, Shen L, Sun G. Squeeze-and-Excitation Networks. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2018; Salt Lake City, USA. https://doi.org/10.1109/CVPR.2018.00745
  9. Aktaş M, Doğan F, Türkoğlu İ. A Hybrid CNN–Transformer Approach for Photovoltaic Cell Defect Classification Using Electroluminescence Imaging. Sensors 2026;26(8):2450. https://doi.org/10.3390/s26082450
  10. Woo S, Park J, Lee JY, Kweon IS. CBAM: Convolutional Block Attention Module. In: European Conference on Computer Vision (ECCV); 2018; Munich, Germany. https://doi.org/10.1007/978-3-030-01234-2_1

Ayrıntılar

Birincil Dil

İngilizce

Konular

Derin Öğrenme

Bölüm

Araştırma Makalesi

Yazarlar

Yayımlanma Tarihi

21 Ekim 2026

Gönderilme Tarihi

7 Temmuz 2026

Kabul Tarihi

14 Ekim 2026

Yayımlandığı Sayı

Yıl 2026 Cilt: 13 Sayı: 1

DOI

https://doi.org/10.54365/adyumbd.1989408

IZ

https://izlik.org/JA52YW36ZF

Kaynak Göster

RIS / Bibtex
APA
Gündoğdu, F., & Çalışan, M. (2026). Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi, 13(1), 1-13. https://doi.org/10.54365/adyumbd.1989408
AMA
1.Gündoğdu F, Çalışan M. Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi. 2026;13(1):1-13. doi:10.54365/adyumbd.1989408
Chicago
Gündoğdu, Fatih, ve Mücahit Çalışan. 2026. “Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset”. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi 13 (1): 1-13. https://doi.org/10.54365/adyumbd.1989408.
EndNote
Gündoğdu F, Çalışan M (01 Ekim 2026) Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi 13 1 1–13.
IEEE
[1]F. Gündoğdu ve M. Çalışan, “Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset”, Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi, c. 13, sy 1, ss. 1–13, Eki. 2026, doi: 10.54365/adyumbd.1989408.
ISNAD
Gündoğdu, Fatih - Çalışan, Mücahit. “Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset”. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi 13/1 (01 Ekim 2026): 1-13. https://doi.org/10.54365/adyumbd.1989408.
JAMA
1.Gündoğdu F, Çalışan M. Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi. 2026;13:1–13.
MLA
Gündoğdu, Fatih, ve Mücahit Çalışan. “Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset”. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi, c. 13, sy 1, Ekim 2026, ss. 1-13, doi:10.54365/adyumbd.1989408.
Vancouver
1.Fatih Gündoğdu, Mücahit Çalışan. Photovoltaic Cell Defect Severity Classification with Squeeze-and-Excitation Enhanced ResNet50 on the ELPV Benchmark Dataset. Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi. 01 Ekim 2026;13(1):1-13. doi:10.54365/adyumbd.1989408