Research on the Defect Detection System of Solar Cell Sheets based on the YOLOv9 Model

Zongyi Zhang; Guoliang Wan

doi:10.54691/0dzkp713

Authors

Zongyi Zhang
Guoliang Wan

DOI:

https://doi.org/10.54691/0dzkp713

Keywords:

Solar Cell Sheets; Typical Defects; Multi-station Visual Detection; YOLOv9m; Model Optimization.

Abstract

A multi-station visual detection system is proposed to address the challenge of detecting various typical defects in solar cell sheets. To improve the detection accuracy of solar cell sheet defects, optimization of the YOLOv9 object detection model is explored. The introduction of the CBAM attention mechanism into the YOLOv9 model's backbone network enhances its ability to extract typical defects. The EIOU loss function replaces the original CIOU loss function, improving both the accuracy of the detection frame and the detection speed. Dynamic Snake Convolution (DSConv) is incorporated into the model to improve its recognition capability for small target defects. The effectiveness of the multi-station visual detection system and the superiority of the optimized model are validated through comprehensive stepwise and comparative experiments across different models.

Downloads

Download data is not yet available.

References

[1] Li Zhenwei, Zhang Shihai,Qu Chongnian, et al. Defect detection for aerospace solar cells based on YOLOX-s algorithm. Acta energiae solaris sinica, Vol.45(2024) No.9,p. 276-284.

[2] WANG J, ZHANG R, ZHENG X. Photovoltaic panel intelligent detection method based on improved faster-RCNN[C]//Proceedings of the IEEE 3rd International Conference on Electronic Technology, Communication and Information. Piscataway:IEEE Press,2023:1565-1569.

[3] LI T, SUN Y. Defect detection of solar panels using improved faster R-CNN[C]//Proceedings of the International Conference on Image, Signal Processing,and Pattern Recognition.SPIE, Vol.127(2024) No.07,p. 605-610.

[4] Peng Xueling, Lin Shanling, Lin Zhixian,et al. Defect detection algorithm of improved YOLOv5s solar cell. Chinese Journal of Liquid Crystals and Displays, Vol.39(2024) No.2,p. 237-247.

[5] CHEN Yafang, LIAO Fei, HUANG Xinyu,et al. Solar cell defect detection using multi-scale YOLOv5. Optics and Precision Engineering,Vol.31(2023) No.12,p. 1804-1815.

[6] JIANG Shang Jun,YI Hui,LI Hong Tao,et al. Defect detection method of solar cells based on transfer learning and ResNet. Acta energiae solaris sinica,Vol.44(2023) No.7,p. 116-121.

[7] Sezen B, Cerasi C C. Solar cell busbars surface defect detection based on deep convolutional neural network[J]. IEEE Latin America Transactions,Vol.21(2023) No.2,p. 242-250.

[8] Yang Shiman,Cao Zheng,Liu Nian Bo,Sun Yanli.Maritime Electro-Optical Image Object Matching Based on Improved YOLOv9.Electronics.Vol.13(2024) No.14,p. 2774-2774.

[9] Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention module[C]//Proceedings of the European conference on computer vision (ECCV). 2018: 3-19.

[10] Du S, Zhang B, Zhang P, et al. An improved bounding box regression loss function based on CIOU loss for multi-scale object detection[C]//2021 IEEE 2nd International Conference on Pattern Recognition and Machine Learning (PRML). IEEE, 2021: 92-98.

[11] Yang Z, Wang X, Li J. EIoU: an improved vehicle detection algorithm based on vehiclenet neural network[C]//Journal of physics: Conference series. IOP Publishing, Vol.1924(2021) No.1,p 012001-012006.

[12] Nascimento M G, Fawcett R, Prisacariu V A. Dsconv: Efficient convolution operator[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2019: 5148-5157.