CSYOLO: a YOLOv8-based PCB defect detection model integrating composite backbone networks and dynamic snake convolution

Abstract

An improved CSYOLOv8 model based on YOLOv8 model is developed specifically for identifying defects in printed circuit board (PCB). Firstly, a composite backbone network is designed to carry out additional feature extraction, which enriches the expression ability of features and enhances the detection accuracy of the model. Secondly, a YOLO-FPN (Feature pyramid network) structure is designed to supplant the original neck network, which enhances the feature fusion ability of the model and improves the detection accuracy of small target objects. Furthermore, to enhance the model’s capability to extract tubular features, dynamic snake convolution is implemented. Finally, MPDIoU loss function is employed to enhance both the convergence rate and the precision of the model. Experiments show that the mAP of the improved model on the PCB defect dataset reaches 96.6%, which is 4.5% higher than that of the YOLOv8 model, and the number of parameters is only 3 256 862, and the average detection speed is 51.8 frames per second, which meets the requirements of detection accuracy and efficiency.

Keywords

printed circuit board (PCB) / deep learning / defect detection / YOLOv8 / multi-scale feature fusion / loss function

Acknowledgement

This work was supported by Natural Science Foundation of Gansu Province (No. 22JR5RA320).

Declaration of conflicting interests

The authors have no conflict of interests related to this publication.

References

[1] ZHOU Y B,YUAN M H,ZHANG J,et al. Review of vision-based defect detection research and its perspectives for printed circuit board. Journal of Manufacturing Systems, 2023, 70: 557-578.

[2] GIRSHICK R. Fast R-CNN//2015 IEEE International Conference on Computer Vision, December 7-13, 2015, Santiago, Chile. New York: IEEE, 2016: 1440-1448.

[3] REN S Q,HE K M,GIRSHICK R,et al. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.

[4] LIU W,ANGUELOV D,ERHAN D,et al. SSD: single shot MultiBox detector//Computer Visio-ECCV 2016. Cham: Springer International Publishing, 2016: 21-37.

[5] REDMON J,DIVVALA S,GIRSHICK R,et al. You only look once: unified, real-time object detection//2016 IEEE Conference on Computer Vision and Pattern Recognition, June 27-30, 2016, Las Vegas, NV, USA. New York: IEEE, 2016: 779-788.

[6] REDMON J,FARHADI A. YOLOv3: an incremental improvement. arXiv:1804.02767.

[7] REDMON J,FARHADI A. YOLO9000: better, faster, stronger//2017 IEEE Conference on Computer Vision and Pattern Recognition, July 21-26, 2017, Honolulu, HI, USA. New York: IEEE, 2017: 6517-6525.

[8] LI C Y,LI L L,JIANG H L,et al. YOLOv6: A Single-Stage object detection framework for industrial applications. arXiv:2209.02976.

[9] WANG C Y,BOCHKOVSKIY A,LIAO H M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 17-24, 2023, Vancouver, BC, Canada. New York: IEEE, 2023: 7464-7475.

[10] DING R W,DAI L H,LI G P,et al. TDD-net: a tiny defect detection network for printed circuit boards. CAAI Transactions on Intelligence Technology, 2019, 4(2): 110-116.

[11] LIN T Y,DOLLÁR P,GIRSHICK R,et al. Feature pyramid networks for object detection//2017 IEEE Conference on Computer Vision and Pattern Recognition, July 21-26, 2017, Honolulu, HI, USA. New York: IEEE, 2017: 936-944.

[12] SHI W,LU Z S,WU W,et al. Single-shot detector with enriched semantics for PCB tiny defect detection. The Journal of Engineering, 2020, 2020(13): 366-372.

[13] WANG L Y,BAI J,LI W J,et al. Research progress of YOLO series target detection algorithms. Computer Engineering and Applications, 2023, 59(14): 15-29.

[14] BIAN B C,CHEN T,WU R J,et al. Improved YOLOv3-based defect detection algorithm for printed circuit board. Journal of Zhejiang University (Engineering Science), 2023, 57(4): 735-743.

[15] ZHANG H,WU C R,ZHANG Z Y,et al. ResNeSt: split-attention networks//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, June 19-20, 2022, New Orleans, LA, USA. New York: IEEE, 2022: 2735-2745.

[16] TANG J L,LIU S B,ZHAO D X,et al. PCB-YOLO: an improved detection algorithm of PCB surface defects based on YOLOv5. Sustainability, 2023, 15(7): 5963.

[17] TAN M X,PANG R M,LE Q V. EfficientDet: scalable and efficient object detection//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 13-19, 2020, Seattle, WA, USA. New York: IEEE, 2020: 10778-10787.

[18] QI Y L,HE Y T,QI X M,et al. Dynamic snake convolution based on topological geometric constraints for tubular structure segmentation//2023 IEEE/CVF International Conference on Computer Vision, October 1-6, 2023, Paris, France. New York: IEEE, 2024: 6047-6056.

[19] MA S L,XU Y. MPDIoU: a loss for efficient and accurate bounding box regression. 2023: arXiv: 2307.07662.

[20] LIU S,QI L,QIN H F,et al. Path aggregation network for instance segmentation//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June. 18-23, 2018, Salt Lake City, UT, USA. New York: IEEE, 2018: 8759-8768.

[21] ZHAO Y A,LV W Y,XU S L,et al. DETRs beat YOLOs on real-time object detection//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 16-22, 2024, Seattle, WA, USA. New York: IEEE, 2024: 16965-16974.

[22] WANG C C,HE W,NIE Y,et al. Gold-YOLO: efficient object detector via gather-and-distribute mechanism//NIPS’23: 37th International Conference on Neural Information Processing Systems, December. 10-16, 2023, New Orleans, LA, USA. Curran Associates Inc, 2023: 51094-51112.

[23] WANG Z Y,LI C,XU H Y,et al. Mamba YOLO: a simple baseline for object detection with state space model. 2024: arXiv: 2406.05835.

[24] TANG S L,HE F,HUANG X L,et al. Online PCB defect detector on a new PCB defect dataset. 2019: arXiv: 1902.06197.