SMG-YOLO: PCB Defect Detection Algorithm Based on Improved Multiscale Fusion Optimization with YOLOv11n

Haojun Liang; Dazhi Yang; Yingqian Zhang

doi:10.62051/ijcsit.v6n2.02

Authors

Haojun Liang
Dazhi Yang
Yingqian Zhang

DOI:

https://doi.org/10.62051/ijcsit.v6n2.02

Keywords:

Printed circuit board (PCB), Defect detection, Deep learning, YOLOv11

Abstract

Printed Circuit Board (PCB) is the substrate of electronic parts, the demand is great, carrying the layout of the circuit components and wires, and its good or bad on the quality of electronic products has important impact! To address the limitations of insufficient detection accuracy and suboptimal real-time performance in current PCB surface defect detection methodologies, this study proposes SMG-YOLO, an enhanced YOLOv11n architecture. The proposed framework integrates three principal innovations: 1) a multi-scale fusion convolution module (MSCB) synergistically combined with C3k2 blocks to optimize feature extraction across spatial resolutions; 2) replacement of the conventional Neck structure with a hierarchical SlimNeck architecture to enhance computational efficiency; and 3) implementation of a Global Context Aggregation mechanism to refine defect localization precision. Experimental validation demonstrates that SMG-YOLO achieves state-of-the-art performance metrics, including a mean average precision (mAP50) of 95.5% and a recall rate of 97.0%, surpassing the baseline YOLOv11 model by 4.3% and 2.0%, respectively. Furthermore, the optimized architecture attains a Computational amount of only 7.7 GFLOPs and a processing speed of 232.8 FPS, satisfying stringent industrial requirements for high-throughput inspection.

Downloads

Download data is not yet available.

References

[1] F. Wang, Z. Yue, et al. Quantitative imaging of printed circuit board (PCB) delamination defects using laser-induced ultrasound scanning imaging [J]. Journal of Applied Physics, 2022, 131(5).

[2] Y. WU, L. ZHAO, Y. YUAN, et al. Research status and outlook of PCB defect detection algorithm based on machine vision [J]. Journal of Instrumentation, 2022, 43(08): 1-17. DOI:10.19650.J2209701.

[3] Y. Zhang, B. Tang. ECARA-YOLO: An improved defect detection algorithm for printed circuit boards [J]. Journal of Harbin University of Commerce (Natural Science Edition), 2025, 41 (01): 43-51. DOI:10.19492/j.cnki.1672-0946.2025.01.003.

[4] Y. XU, J. Xiao, Y. Wang, et al. A research review on PCB surface defect detection based on machine vision [J]. Microelectronics and Computers, 2025, 42 (04): 1-15. DOI:10.19304/J.ISSN1000-7180.2024.0186.

[5] R. Lu, A. Wu, T. Zhang, et al. A review of automatic optical (vision) inspection technology and its application in defect detection [J]. Journal of Optics, 2018, 38 (08): 23-58.

[6] Support vector machines: theory and applications [M]. Springer Science & Business Media, 2005.

[7] Zou J, Han Y, So S S. Overview of artificial neural networks [J]. Artificial neural networks: methods and applications, 2009: 14-22.

[8] Garg P. A Comparison between Memetic algorithm and Genetic algorithm for the cryptanalysis of Simplified Data Encryption Standard algorithm [J]. arXiv preprint arXiv:1004.0574, 2010.

[9] Kotsiantis S B. Decision trees: a recent overview [J]. Artificial Intelligence Review, 2013, 39: 261-283.

[10] Liao, X.; Lv, S.; Li, D.; Luo, Y.; Zhu, Z.; Jiang, C. Yolov4-mn3 for pcb surface defect detection. Appl. Sci. 2021, 11, 11701.

[11] Du, B.; Wan, F.; Lei, G.; Xu, L.; Xu, C.; Xiong, Y. YOLO-MBBi: PCB Surface Defect Detection Method Based on Enhanced YOLOv5. Electronics 2023, 12, 2821.

[12] Ng D, Chen Y, Tian B, et al. Convmixer: Feature interactive convolution with curriculum learning for small footprint and noisy far-field keyword spotting [C]//ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2022: 3603-3607.

[13] Chen J, Mai H S, Luo L, et al. Effective feature fusion network in BIFPN for small object detection [C]//2021 IEEE international conference on image processing (ICIP). IEEE, 2021: 699-703.

[14] Xiao, G., Hou, S. & Zhou, H. PCB defect detection algorithm based on CDI-YOLO. Sci Rep 14, 7351 (2024).

[15] 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.

[16] 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.

[17] 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.

[18] Mo C, Hu Z, Wang J, et al. SGT-YOLO: A Lightweight Method for PCB Defect Detection [J]. IEEE Transactions on Instrumentation and Measurement, 2025.

[19] Jocher G, Stoken A, Borovec J, et al. ultralytics/yolov5: v3. 0 [J]. Zenodo, 2020.

[20] Tan M, Le Q. Efficientnetv2: Smaller models and faster training [C]//International conference on machine learning. PMLR, 2021: 10096-10106.

[21] Zhuang J, Qin Z, Yu H, et al. Task-specific context decoupling for object detection [J]. arXiv preprint arXiv:2303.01047, 2023.

[22] Khanam, R., & Hussain, M. (2024). Yolov11: An overview of the key architectural enhancements. arxiv preprint arxiv:2410.17725.

[23] Li H, Li J, Wei H, et al. Slim-neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles [J]. arXiv preprint arXiv: 2206.02424, 2022, 10.

[24] Vinyals O, Blundell C, Lillicrap T, et al. Matching networks for one shot learning [J]. Advances in neural information processing systems, 2016, 29.

[25] Rahman M M, Munir M, Marculescu R. Emcad: Efficient multi-scale convolutional attention decoding for medical image segmentation [C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2024: 11769-11779.

[26] Liu Y, Shao Z, Hoffmann N. Global attention mechanism: Retain information to enhance channel-spatial interactions [J]. arXiv preprint arXiv:2112.05561, 2021.

[27] 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.