PCB Component Recognition based on Improved YOLOv8s for Lightweight Detection

Si Shen; Juan Wang; Fan Yang

doi:10.6919/ICJE.202504_11(4).0002

Authors

Si Shen
Juan Wang
Fan Yang

DOI:

https://doi.org/10.6919/ICJE.202504_11(4).0002

Keywords:

YOLOv8s; Small Object Detection; Components; Lightweight; Multi-Scale Features.

Abstract

To address the challenges posed by the diverse types of printed circuit board (PCB) components, the low detection accuracy for small components, and high model complexity, this paper proposes the GS-YOLOv8s algorithm. By introducing lightweight Ghost convolution, the computational costs are reduced. A dynamic lightweight C2f-GDC module is designed by incorporating the Ghost Module and DynamicConv into the C2f module of the Backbone, effectively minimizing redundant information within the model. In the Neck network, the SE module is integrated into the bottleneck layer of the C2f module, forming the C2f-SE module, which enhances the network's capability to extract features of small objects. Additionally, a multi-scale feature detection layer is incorporated, leveraging fine-grained information from shallow layers to substantially improve the detection accuracy of small components. Experimental validation conducted on a PCB component dataset demonstrates that GS-YOLOv8s achieves a detection accuracy of 95.6%, a recall rate of 94.6%, and an mAP@0.5 of 97.4%, outperforming YOLOv8s by 0.7%, 1.4%, and 0.5%, respectively. Notably, the model's parameter count is reduced by 30%. The results affirm that the proposed algorithm effectively enhances the detection capability for small objects in complex scenarios, while maintaining high detection accuracy and computational efficiency.

Downloads

Download data is not yet available.

References

[1] Chen J, Bao E, Pan J. Classification and positioning of circuit board components based on improved YOLOv5[J]. Procedia Computer Science, 2022, 208: 613-626.

[2] Mahalingam G, Gay K M, Ricanek K. Pcb-metal: A pcb image dataset for advanced computer vision machine learning component analysis[C]//2019 16th International Conference on Machine Vision Applications (MVA). IEEE, 2019: 1-5.

[3] Sharma H, Kumar H. A computer vision-based system for real-time component identification from waste printed circuit boards[J]. Journal of Environmental Management, 2024, 351: 119779.

[4] Luo S, Wan F, Lei G, et al. EC-YOLO: Improved YOLOv7 Model for PCB Electronic Component Detection[J]. Sensors, 2024, 24(13): 4363.

[5] Ling Q, Isa N A M, Asaari M S M. Precise detection for dense PCB components based on modified YOLOv8[J]. IEEE Access, 2023.

[6] Wang C Y, Bochkovskiy A, Liao H Y M. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2023: 7464-7475.

[7] Wang G, Chen Y, An P, et al. UAV-YOLOv8: A small-object-detection model based on improved YOLOv8 for UAV aerial photography scenarios[J]. Sensors, 2023, 23(16): 7190.

[8] Zhu Hongyan, Li Zeping, Zhao Yong, et al. PCB defect detection algorithm based on multi-scale fusion and deformable convolution[J]. Computer Engineering and Design, 2022, 43(08): 2188-2196.

[9] Han K, Wang Y, Tian Q, et al. Ghostnet: More features from cheap operations[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020: 1580-1589.

[10] Wu L, Zhang L, Zhou Q. Printed circuit board quality detection method integrating lightweight network and dual attention mechanism[J]. IEEE Access, 2022, 10: 87617-87629.

[11] Chen Y, Dai X, Liu M, et al. Dynamic convolution: Attention over convolution kernels[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020: 11030-11039.

[12] Wang B, Hua J, Xia L, et al. A defect detection method for Akidzuki pears based on computer vision and deep learning[J]. Postharvest Biology and Technology, 2024, 218: 113157.

[13] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.

[14] Li H, Gu Z, He D, et al. A lightweight improved YOLOv5s model and its deployment for detecting pitaya fruits in daytime and nighttime light-supplement environments[J]. Computers and Electronics in Agriculture, 2024, 220: 108914.

[15] Gui Xiangquan, Qing Qingsong, Kong Lingwang. Small object detection algorithm based on improved YOLOv5s[J]. Computer Engineering and Design, 2024,45(04):1134-1140.

[16] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.

[17] Hou Q, Zhou D, Feng J. Coordinate attention for efficient mobile network design[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2021: 13713-13722.

[18] Wang Q, Wu B, Zhu P, et al. ECA-Net: Efficient channel attention for deep convolutional neural networks[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020: 11534-11542.

[19] 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.

[20] Ren S, He K, Girshick R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2016, 39(6): 1137-1149.

[21] Liu W, Anguelov D, Erhan D, et al. Ssd: Single shot multibox detector[C]//Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14. Springer International Publishing, 2016: 21-37.

[22] Ross T Y, Dollár G. Focal loss for dense object detection[C]//proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 2980-2988.

[23] Adarsh P, Rathi P, Kumar M. YOLO v3-Tiny: Object Detection and Recognition using one stage improved model[C]//2020 6th international conference on advanced computing and communication systems (ICACCS). IEEE, 2020: 687-694.

[24] Wang C Y, Bochkovskiy A, Liao H Y M. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2023: 7464-7475.