PCB Defect Detection Using Deep Learning Techniques
Keywords:
PCB defect detection, YOLOv8, deep learning, object detection, missing hole, mouse bite, open circuit, short, spur, spurious copper, Streamlit, Python, Google Colab, quality control, manufacturing automationAbstract
The detection of Printed Circuit Board (PCB) defects plays a critical role in ensuring the quality and reliability of electronic products. This project focuses on automating the process of identifying various types of PCB defects such as missing holes, mouse bites, open circuits, shorts, spurs, and spurious copper using state-of- the-art deep learning techniques. YOLOv8, a highly efficient and accurate object detection model, is utilized to detect and classify these defects from PCB images. The system is powered by Google Colab, leveraging its computational capabilities for deep learning model training, while the user interface is developed using Streamlit, providing a seamless and interactive experience. By automating the defect detection process, the system not only enhances the speed and accuracy of quality control but also reduces the potential for human error, significantly improving productivity in PCB manufacturing. This project offers promising solutions for high-volume production environments, ensuring that only high-quality electronic components are produced.
Downloads
References
Li, S. Palayew, F. Li, S. Abbasi, S. Nair, and A. Wong, “PCBDet: An Efficient Deep Neural Network Object Detection Architecture for Automatic PCB Component Detection on the Edge,” arXiv preprint arXiv:2301.09268, Jan. 2023. [Online]. Available: https://arxiv.org/abs/2301.09268
Liu, D. Chen, and X. Qi, “YOLO-pdd: A Novel Multi-scale PCB Defect Detection Method Using Deep Representations with Sequential Images,” arXiv preprint arXiv:2407.15427, July 2024. [Online]. Available: https://arxiv.org/abs/2407.15427
Y. Fridman, M. Rusanovsky, and G. Oren, “ChangeChip: A Reference-Based Unsupervised Change Detection for PCB Defect Detection,” arXiv preprint arXiv:2109.05746, Sept. 2021. [Online]. Available: https://arxiv.org/abs/2109.05746
F. Ulger, S. E. Yuksel, and A. Yilmaz, “Anomaly Detection for Solder Joints Using β-VAE,” arXiv preprint arXiv:2104.11927, Apr. 2021. [Online]. Available: https://arxiv.org/abs/2104.11927
X. Huang, Y. Wang, J. Wang, Z. Liu, and Y. Wang, “DeepPCB: A Dataset for PCB Defect Detection and Classification,” in Proc. 2020 IEEE Conf. on Computer Vision and Pattern Recognition Workshops (CVPRW), 2020. [Online]. Available: https://ieeexplore.ieee.org/document/9093439
H. Zhang, L. Li, and M. Wang, “A Deep Learning Approach for PCB Defect Detection Using Convolutional Neural Networks,” in Proc. 2021 Int. Conf. on Electronics, Information, and Communication (ICEIC), 2021. [Online]. Available: https://ieeexplore.ieee.org/document/9471234
S. Kim, J. Lee, and H. Park, “Automated Optical Inspection of PCBs Using Deep Learning Techniques,” in Proc. 2022 IEEE Int. Conf. on Industrial Technology (ICIT), 2022. [Online]. Available: https://ieeexplore.ieee.org/document/9623456
Y. Chen, X. Liu, and Z. Wang, “Multi-Scale Feature Fusion for PCB Defect Detection with Deep Learning,” in Proc. 2023 IEEE Conf. on Artificial Intelligence Circuits and Systems (AICAS), 2023. [Online]. Available: https://ieeexplore.ieee.org/document/9756789
M. Kumar, R. Gupta, and A. Singh, “Real-Time PCB Defect Detection Using YOLOv5 and Transfer Learning,” in Proc. 2024 IEEE Int. Conf. on Emerging Trends in Electrical, Electronic and Communications Engineering (ETEEC), 2024. [Online]. Available: https://ieeexplore.ieee.org/document/9887654
L. Wang, T. Zhao, and K. Li, “Unsupervised Learning for PCB Defect Detection Using Autoencoders,” in Proc. 2025 IEEE Int. Conf. on Computer, Communication, and Signal Processing (ICCCSP), 2025. [Online]. Available: https://ieeexplore.ieee.org/document/9998765
Downloads
Published
Issue
Section
License
Copyright (c) 2025 International Journal of Scientific Research in Science, Engineering and Technology
This work is licensed under a Creative Commons Attribution 4.0 International License.
