Printed Circuit Boards (PCBs)為電子產品中關鍵的零件之一，為了確保電子產品長期使用和穩定性能，PCB瑕疵檢測為關鍵一環。現有PCB瑕疵檢測深度學習模型在精度上取得良好成績，然而卻未考量到模型大小以及即時性。因此，本研究旨在基於YOLOv5模型進行改進，提出了精準且快速的輕量化瑕疵檢測模型CCG-YOLO。本研究引入C3HB模組改良骨幹網路，通過空間交互作用強化特徵提取能力。在特徵融合網路中引入GSConv輕量卷積結構，在保持模型性能下大幅縮減模型參數。為了更進一步提升模型性能，本研究引入基於ConvNeXt網路之CNeB模組替換Neck中C3模組，CNeB在高效殘差網路設計下，不僅提高了模型檢測精度，更能減少模型參數量。綜合上述操作，本研究提出模型在TDD-Net公開數據集上，模型準確度mAP@0.5達到99.5%以及mAP@0.5:0.95達到88.75%。和原YOLOv5s算法相比在mAP@0.5:0.95上升4.24%，模型Size下降1MB，參數量減少0.472M，計算複雜度下降了0.6GFLOPs，即時推理速度達到120 FPS。實驗結果表明，本研究提出模型在PCB瑕疵檢測上精度高、速度快且模型體積小，能輕易佈署至低階設備同時滿足工業瑕疵檢測上即時性需求。

Printed Circuit Boards (PCBs) are one of the critical components in electronic products. To ensure the long-term use and stable performance of these products, PCB defect detection is a key process. While the existing deep learning models for PCB defect detection have achieved good results in terms of accuracy, they have not taken into account the model size and real-time processing ability. Therefore, this study proposes a precise and fast lightweight defect detection model based on an improved YOLOv5 model called CCG-YOLO. Firstly, the backbone network is improved by adding a C3HB module, which enhances the feature extraction ability through spatial interaction capability. Secondly, this study introduces a lightweight convolution structure, GSConv, in the feature fusion network, significantly reducing model parameters while maintaining performance. Lastly, to further improve model performance, a CNeB module based on ConvNeXt network is introduced to replace the C3 module in the Neck. Designed with efficient residual networking, CNeB not only improves model detection accuracy but also reduces the number of model parameters. By combining these operations, CCG-YOLO achieved a mean average precision (mAP@0.5) of 99.5% and 88.75% in mAP@0.5:0.95 on the TDD-Net public dataset. Compared with the original YOLOv5s algorithm, the improvement in mAP@0.5:0.95 was 4.24%, the model size was reduced by 1MB, the number of parameters was decreased by 0.472M, computational complexity was reduced by 0.6GFLOPs, and the real-time inference speed reached 120 FPS. The experimental results show that the model proposed in this study is highly accurate, fast, and has a small model size for PCB defect detection. Furthermore, CCG-YOLO can easily be deployed to low-end devices and meet the real-time requirements of industrial defect detection.

Adibhatla, V. A., Chih, H. C., Hsu, C. C., Cheng, J., Abbod, M. F., & Shieh, J. S. (2021). Applying deep learning to defect detection in printed circuit boards via a newest model of you-only-look-once.
Adibhatla, V. A., Chih, H. C., Hsu, C. C., Cheng, J., Abbod, M. F., & Shieh, J. S. (2020). Defect detection in printed circuit boards using you-only-look-once convolutional neural networks. Electronics, 9(9), 1547.
An, Q., Wang, K., Li, Z., Song, C., Tang, X., & Song, J. (2022). Real-Time Monitoring Method of Strawberry Fruit Growth State Based on YOLO Improved Model. IEEE Access, 10, 124363-124372.
Bochkovskiy, A., Wang, C. Y., & Liao, H. Y. M. (2020). Yolov4: Optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934.
Chen, W., Huang, Z., Mu, Q., & Sun, Y. (2022). PCB Defect Detection Method Based on Transformer-YOLO. IEEE Access, 10, 129480-129489.
Chollet, F. (2017). Xception: Deep learning with depthwise separable convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1251-1258).
Ding, R., Dai, L., Li, G., & Liu, H. (2019). TDD‐net: a tiny defect detection network for printed circuit boards. CAAI Transactions on Intelligence Technology, 4(2), 110-116.
Girshick, R. (2015). Fast r-cnn. In Proceedings of the IEEE international conference on computer vision (pp. 1440-1448).
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 580-587).
Han, K., Wang, Y., Tian, Q., Guo, J., Xu, C., & Xu, C. (2020). Ghostnet: More features from cheap operations. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 1580-1589).
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
Hu, B., & Wang, J. (2020). Detection of PCB surface defects with improved faster-RCNN and feature pyramid network. IEEE Access, 8, 108335-108345.
Huang, W., Wei, P., Zhang, M., & Liu, H. (2020). HRIPCB: a challenging dataset for PCB defects detection and classification. The Journal of Engineering, 2020(13), 303-309.
Jocher, G., Stoken, A., Borovec, J., Changyu, L., Hogan, A., Diaconu, L., ... & Reñé Claramunt, E. (2020). ultralytics/yolov5: v3. 0. Zenodo.
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017). Imagenet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84-90.
LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324.
Li, H., Li, J., Wei, H., Liu, Z., Zhan, Z., & Ren, Q. (2022). Slim-neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles. arXiv preprint arXiv:2206.02424.
Li, Z., & Yang, Q. (2011, October). System design for PCB defects detection based on AOI technology. In 2011 4th International Congress on Image and Signal Processing (Vol. 4, pp. 1988-1991). IEEE.
Liao, X., Lv, S., Li, D., Luo, Y., Zhu, Z., & Jiang, C. (2021). Yolov4-mn3 for pcb surface defect detection. Applied Sciences, 11(24), 11701.
Lim, J., Lim, J., Baskaran, V. M., & Wang, X. (2023). A deep context learning based PCB defect detection model with anomalous trend alarming system. Results in Engineering, 17, 100968.
Lin, T. Y., Dollár, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017). Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2117-2125).
Ling, Q., & Isa, N. A. M. (2023). Printed Circuit Board Defect Detection Methods Based on Image Processing, Machine Learning and Deep Learning: A Survey. IEEE Access.
Liu, F., & Shen, Y. (2022, November). A modified-yolov5s model for defect detection of printed circuit board. In 2022 China Automation Congress (CAC) (pp. 351-356). IEEE.
Liu, G., Hu, Y., Chen, Z., Guo, J., & Ni, P. (2023). Lightweight object detection algorithm for robots with improved YOLOv5. Engineering Applications of Artificial Intelligence, 123, 106217.
Liu, S., Qi, L., Qin, H., Shi, J., & Jia, J. (2018). Path aggregation network for instance segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 8759-8768).
Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C. (2016). Ssd: Single shot multibox detector. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14 (pp. 21-37). Springer International Publishing.
Liu, Z., Mao, H., Wu, C. Y., Feichtenhofer, C., Darrell, T., & Xie, S. (2022). A convnet for the 2020s. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 11976-11986).
Rao, Y., Zhao, W., Tang, Y., Zhou, J., Lim, S. N., & Lu, J. (2022). Hornet: Efficient high-order spatial interactions with recursive gated convolutions. Advances in Neural Information Processing Systems, 35, 10353-10366.
Redmon, J., & Farhadi, A. (2017). YOLO9000: better, faster, stronger. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 7263-7271).
Redmon, J., & Farhadi, A. (2018). Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767.
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 779-788).
Shi, W., Lu, Z., Wu, W., & Liu, H. (2020). Single‐shot detector with enriched semantics for PCB tiny defect detection. The Journal of Engineering, 2020(13), 366-372.
Shu, Y., Li, B., & Lin, H. (2021). Quality safety monitoring of LED chips using deep learning-based vision inspection methods. Measurement, 168, 108123.
Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
Tang, J., Liu, S., Zhao, D., Tang, L., Zou, W., & Zheng, B. (2023). PCB-YOLO: An Improved Detection Algorithm of PCB Surface Defects Based on YOLOv5. Sustainability, 15(7), 5963.
Wang, C. Y., Liao, H. Y. M., Wu, Y. H., Chen, P. Y., Hsieh, J. W., & Yeh, I. H. (2020). CSPNet: A new backbone that can enhance learning capability of CNN. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops (pp. 390-391).
Wang, J., Fu, P., & Gao, R. X. (2019). Machine vision intelligence for product defect inspection based on deep learning and Hough transform. Journal of Manufacturing Systems, 51, 52-60.
Wu, F., Zhang, X., Kuan, Y., & He, Z. (2008, June). An AOI algorithm for PCB based on feature extraction. In 2008 7th World Congress on Intelligent Control and Automation (pp. 240-247). IEEE.
Wu, H., Gao, W., & Xu, X. (2019). Solder joint recognition using mask R-CNN method. IEEE Transactions on Components, Packaging and Manufacturing Technology, 10(3), 525-530.
Zhang, L., Jin, Y., Yang, X., Li, X., Duan, X., Sun, Y., & Liu, H. (2018). Convolutional neural network‐based multi‐label classification of PCB defects. The Journal of Engineering, 2018(16), 1612-1616.
Zheng, Y., Zhang, Y., Qian, L., Zhang, X., Diao, S., Liu, X., ... & Huang, H. (2023). A lightweight ship target detection model based on improved YOLOv5s algorithm. Plos one, 18(4), e0283932.