Yolo Based Defects Detection Algorithm for EL in PV Modules with Focal and Efficient IoU Loss
Source abstract
Abstract
Considering the defect detection issues in electroluminescence (EL) of photovoltaic (PV) cell systems, lots of factors result in performance degradation, including defect diversity, data imbalance, scale difference, etc. Focal-EIoU loss, an effective defect detection solution for EL, is proposed based on the improved YOLOv5. Firstly, by analyzing the detection background and scale characteristics of EL defects, a binary classification is carried out in the system. Subsequently, a cascade detection network based on YOLOv5 is designed to further extract features from the binary-classified defects. The defect localization and classification are achieved in this way. To address the problem of imbalanced defect samples, a loss function is designed based on EIoU and Focal-F1 Loss. Experimental results are illustrated to show the effectiveness. Compared with the existing CNN-based deep learning approaches, the proposed focal loss calculation-based method can effectively improve the performance of handling sample imbalance. Moreover, in the detection of 12 types of defects, the Yolov5 algorithms can always obtain higher MAP (mean average precision) even with different parameter levels (Yolov5m: 0.791 vs. 0.857, Yolov5l: 0.798 vs. 0.862, Yolov5x: 0.802 vs. 0.867, Yolov5s: 0.793 vs. 0.865).
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Photovoltaic System Optimization Techniques · Currency Recognition and Detection
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Automation, Southeast University, Nanjing 210096, China
Shenzhen Research Institute, Southeast University, Shenzhen 518000, China
Funding
Shenzhen Science and Technology Program · Guangdong Basic and Applied Basic Research Foundation
References (47)
Show source-provided references
1. Hu, X., Tang, W., Bi, J., Chen, S., and Yan, W. (2021, January 26–28). CNN-based model for cell extraction from PV modules with EL images for PV defects detection. Proceedings of the Chinese Control Conference, Shanghai, China.10.23919/CCC52363.2021.9549808 ↗
2. Dou, Z. (2013). Research of Online Defects Detection for Solar Panel Base on the EL Image. [Master’s Thesis, Zhejiang Sci-Tech University].
3. Wang, Z. (2014). Research on Defect Detection System for Solar Cells. [Master’s Thesis, Hebei University of Technology].
4. Solar cells surface defects detection based on deep learning
5. Defect recognition for radiographic image based on deep learning network
6. Chen, T., Ding, S., Chen, H., and Chen, C. (2022, January 1–3). High accuracy detection strategy for EL defects in PV modules based on machine learning. Proceedings of the Asia-Pacific Conference on Intelligent Robot Systems, Tianjin, China.10.1109/ACIRS55390.2022.9845525 ↗
7. Yang, J., and Guo, J. (2011, January 4–5). Image texture feature extraction method based on regional average binary gray level difference co-occurrence matrix. Proceedings of the International Conference on Virtual Reality and Visualization, IEEE Computer Society, Beijing, China.10.1109/ICVRV.2011.20 ↗
8. Defect detection of polycrystalline solar wafers using local binary mean10.1007/s00170-015-7498-z ↗
9. Automatic weld defect detection method based on Kalman filtering for real-time radiographic inspection of spiral pipe10.1016/j.ndteint.2015.01.002 ↗
10. Defect detection of small hole inner surface based on multi-focus image fusion
11. Frome, A., Singer, Y., Sha, F., and Malik, J. (2007, January 14–21). Learning globally-consistent local distance functions for shape-based image retrieval and classification. Proceedings of the IEEE 11th International Conference on Computer Vision, Rio de Janeiro, Brazil.10.1109/ICCV.2007.4408839 ↗
12. An image mosaic method for defect inspection of steel rotary parts10.1007/s10921-016-0375-3 ↗
13. Han, J. (2020). Research on Classification and Detection of Solar Cells EL Defects under Class Imbalance Situation. [Master’s Thesis, Hebei University of Technology].
14. Bartler, A., Mauch, L., Yang, B., Reuter, M., and Stoicescu, L. (2018, January 3–7). Automated detection of solar cell defects with deep learning. Proceedings of the 2018 26th European Signal Processing Conference (EUSIPCO), Rome, Italy.10.23919/EUSIPCO.2018.8553025 ↗
15. Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv.
16. Research on detection technology of photovoltaic cells based on GoogLeNet and EL
17. Automatic classification of defective photovoltaic module cells in electroluminescence images10.1016/j.solener.2019.02.067 ↗
18. Kaur, A., Kukreja, V., Kumar, M., Choudhary, A., and Sharma, R. (2024, January 24–26). A Fine-tuned Deep Learning-based VGG16 Model for Cotton Leaf Disease Classification. Proceedings of the 2024 5th International Conference for Emerging Technology (INCET), Belgaum, India.10.1109/INCET61516.2024.10593164 ↗
19. Deep learning -based automatic defect identification of the photovoltaic module using electroluminescence images10.1016/j.solener.2020.03.049 ↗
20. BAF-detector: An efficient CNN-based detector for photovoltaic cell defect detection10.1109/TIE.2021.3070507 ↗
21. Dong, M. (2018). Research on Yarn-dyed Fabric Defect Detection and Classification based on Convolutional Neural Network. [Master’s Thesis, Xi’an Polytechnic University].
22. Defect detection of continuous casting slabs based on deep learning
23. Faster R-CNN: Towards teal-Time object detection with region proposal networks10.1109/TPAMI.2016.2577031 ↗
24. Yang, Q., Ma, S., Guo, D., Wang, P., Lin, M., and Hu, Y. (2023). A small object detection method for oil leakage defects in substations based on improved faster-RCNN. Sensors, 23.10.3390/s23177390 ↗
25. Detection for pin defects of overhead lines by UAV patrol image based on improved Faster-RCNN
26. Tang, Y., Han, J., and Wei, W. (2018). Research on part recognition and detection of transmission line in deep learning. Electron. Meas. Technol., 60–65.
27. Dai, X., Chen, H., and Zhu, C. (2020). Surface defect detection and realization of mental workpiece based on improved Faster-RCNN. Surf. Technol., 362–371.
28. Guo, M., and Xu, H. (2019). Hot spot defect detection based on infrared thermal image and faster-RCNN. Comput. Syst. Appl., 265–270.
29. Zhu, C., and Yang, Y. (2020). Online detection algorithm of automobile wheel surface defects based on improved Faster-RCNN model. Surf. Technol., 358–365.
30. Guo, T., Yang, H., and Shi, L. (2019). Self-explosion defect identification of insulator based on faster RCNN. Insul. Surge Arresters, 183–189.
31. Xu, X. (2019). Solar Cell Defect Detection Based on Deep Learning. [Master’s Thesis, North University of China].
32. Tian, X., Cheng, Y., and Chan, G. (2019). Detect and recognize hidden cracks in silicon chips on deep learning SSD algorithm. Mach. Tool Hydraul., 36–41.
33. Focal loss for dense object detection
34. Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016, January 27–30). You only look once: Unified, real-time object detection. Proceedings of the IEEE International Conference on Computer Vision, Las Vegas, NV, USA.10.1109/CVPR.2016.91 ↗
35. Redmon, J., and Farhadi, A. (2017, January 21–26). YOLO9000: Better, faster, stronger. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.690 ↗
36. Redmon, J., and Farhadi, A. (2018, January 18–23). YOLOv3: An incremental improvement. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.
37. Bochkovskiy, A., Wang, C., and Mark-Liao, H. (2020, January 13–19). YOLOv4: Optimal Speed and Accuracy of Object Detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.
38. A robust single-sensor MPPT strategy for a shaded photovoltaic-battery system10.32604/csse.2021.015029 ↗
39. A novel framework for intelligent detection for module defects of PV plant combining the visible and infra-red images10.1016/j.solener.2022.03.018 ↗
40. Automatic fabric defect detection based on an improved YOLOv510.1155/2021/7321394 ↗
41. Zhu, X. (2021, January 23–25). Design of barcode recognition system based on YOLOV5. Proceedings of the 2021 3rd International Conference on Computer Modeling, Changsha, China.
42. Jin, S., and Sun, L. (2021, January 22–24). Application of enhanced feature fusion applied to YOLOv5 for ship detection. Proceedings of the 33rd Chinese Control and Decision Conference, Kunming, China.10.1109/CCDC52312.2021.9602100 ↗
43. Activation functions in deep learning: A comprehensive survey and benchmark10.1016/j.neucom.2022.06.111 ↗
44. Multi-spatial pyramid feature and optimizing focal loss function for object detection10.1109/TIV.2023.3282996 ↗
45. Wang, Y., Tian, Y., and Cheng, J. (2023, January 20–22). An improved YOLOv7 method for vehicle detection in traffic scenes. Proceedings of the 2023 35th Chinese Control and Decision Conference (CCDC), Yichang, China.10.1109/CCDC58219.2023.10326522 ↗
46. Li, C., Li, L., Jiang, H., Weng, K., Geng, Y., Li, L., Ke, Z., Li, Q., Cheng, M., and Nie, W. (2022). YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications. arXiv.
47. Wang, C., Bochkovskiy, A., and Liao, H. (2023, January 17–24). YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors. Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.10.1109/CVPR52729.2023.00721 ↗