Article Details

Abstract

To address the various defects (such as open circuits, short circuits, missing vias, burrs, and copper leakage) that are prone to occur during the production of PCBs with increasingly complex manufacturing processes, as well as the problems of low detection accuracy for small defects, complex models, and high computational cost, this study proposes an improved detection algorithm based on YOLOv11, named DEW-YOLO. The algorithm utilizes multi-scale features to provide richer and more stable representations and employs an EMA attention module to handle complex background interference and small target recognition. Finally, the CIoU loss function is replaced with the WIoU loss function to capture the feature information more fully, thereby effectively enhancing the overall detection performance and improving the detection of low-quality small targets. Testing shows that the DEW-YOLO model achieves an mAP of 93.8% in PCB defect detection, a 4.9% improvement over YOLOv11, achieving the best performance among similar algorithms.

Keywords

PCB; YOLOv11; EMA attention; target detection

References

1. [1] Zhou Xianyong, Zhu Junhui, Wang Yuan, etc PCB bare board defect detection based on improved YOLOv7 algorithm [J]. Radio Engineering, 2023, 53 (12): 2791-2797
2. [2] Liu Shuang, Lv Junliang, Qin Yuhang, etc Improved Algorithm for PCB Defect Detection Based on YOLOv8 [J]. Journal of Jilin University (Science Edition), 2026, 64 (2): 344-350
3. [3] Song Xudong, Sun Yongbo, Song Liang, etc Research on Surface Defect Detection Method of Circuit Board Based on Object Detection [J]. Journal of Dalian Jiaotong University, 2025, 46 (2): 113-120
4. [4] REN S Q，HE K M，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，2017，39（6）：1137-1149.
5. [5] HE K M，GKIOXARI G，DOLLÁR P，et al. Mask R-CNN [C]//Proceedings of the IEEE International Conference on Computer Vision.Los Alamitos：IEEE Computer Society Press，2017：2961-2969.
6. [6] REDMON J，DIVVALA S，GIRSHICK R，et al. You Only Look Once: Unified, Real-Time Object Detection [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition（CVPR）.Los Alamitos：IEEE Computer Society，2016：779-788.
7. [7] Wang Caixia, Guo Xinpeng, Liu Peng Improved YOLOv8n circuit board defect detection algorithm [J]. Journal of Electronic Measurement and Instrumentation, 2025, 39 (8): 65-78
8. [8] LIN T Y，DOLLAR P，GIRSHICK R，et al. Feature Pyramid Networks for Object Detection [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition（CVPR）.Los Alamitos：IEEE，2017：936-944.
9. [9] XU S，ZHENG S C，XU W，et al. HCF-Net: Hierarchical Context Fusion Network for Infrared Small Object Detection [C]//2024 IEEE International Conference on Multimedia and Expo （ICME）. IEEE，2024：1-6.
10. [10] OUYANG D，HE S，ZHAN J，et al. Efficient Multi-Scale Attention Module with Cross-Spatial Learning [PP/OL].arXiv.（2023-06-06）.https://arxiv.org/abs/2305.13563v1.
11. [11] WANG J，XU C，YANG W，et al. A Normalized Gaussian Wasserstein Distance for Tiny Object Detection [J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2021，45（1）：630-643.
12. [12] Wu Jigang, Cheng Yuan, Shao Jun, etc Research on PCB Defect Detection Based on Improved YOLOv4 Algorithm [J]. Journal of Instrumentation, 2021, 42 (10): 171-178
13. [13] Huang Wenjie, Luo Weiping, Chen Zhennan, etc Research on Lightweight PCB Defect Detection Algorithm Based on YOLO11 [J]. Journal of Guangxi Normal University (Natural Science Edition), 2026, 44 (1): 56-67
14. [14] Zhao Hongchao Research on PCB Component Defect Detection System Based on Improved Faster R-CNN [D]. Chongqing: Chongqing University of Science and Technology, 2023
15. [15] TONG Z，CHEN Y，XU Z，et al. Wise-IoU: Bounding Box Regression Loss with Dynamic Focusing Mechanism [PP/OL].arXiv.（2023-04-08）.https://arxiv.org/pdf/2301.10051v2.
16. [16] ZHANG K，WANG Y，SHI S，et al. SF-YOLO：Designed Based on Tiny Feature for PCB Surface Defect Detection and Deployment in Embedded Systems [J].Signal, Image and Video Processing，2025，19（3）：263-263.
17. [17] WANG J，MA L，LI Z，et al. SSHP-YOLO: A High Precision Printed Circuit Board （PCB） Defect Detection Algorithm with a Small Sample [J].Electronics，2025，14（2）：217-217.
18. [18] Hu Chao, Du Yulong, Dong Tianle, etc PCB defect detection algorithm based on YOLOv8n PSA [J]. Laboratory Testing, 2025, 3 (21): 51-53
19. [19] Peng Zaihuan, Chang Guangchao, Ren Chuancheng PCB surface defect detection algorithm based on DLDE-YOLOv10n [J]. Journal of Ningxia Normal University, 2025, 46 (4): 60-70

About the authors

Xu Sen (1975-), male, Han ethnicity, from Jinzhou, Liaoning, lecturer, doctoral degree, research interests include network security and cryptography, security protocol design and analysis and wireless network protocols; Xiao Bingyao (2000-), female, Han ethnicity, from Anshan, Liaoning, postgraduate student, research interests include deep learning and machine learning; Corresponding author: Zhao Yang (1974-), male, Hui ethnicity, from Shenyang, Liaoning, lecturer, doctoral degree, research interests include deep learning, pattern recognition and image processing.