YOLOv8 tabanlı PCB Hata Tespit ve Sınıflandırma Sistemi

Öz

Baskı Devre Kartlarının yüzey muayenesi, üretim sırasında oluşabilecek küçük hataların bile ciddi maliyetlere yol açması nedeniyle en önemli kalite kontrol süreçlerinden biridir. Bu çalışmada baskı devrelerde (BD) sık karşılaşılan altı hatanın tespiti ve sınıflandırılması için YOLOv8 tabanlı bir sistem geliştirilmiştir. Doğruluk, hız ve aynı anda birden fazla hatayı tespit edebilme yeteneği nedeni ile önerilen yöntem, diğer BD hata tespit yöntemlerine kıyasla, üretim bantlarında kullanıma daha uygundur. Sistem aynı zamanda hedefe yönelik denetim için özelleştirilebilir hata seçim olanağı da sunmaktadır. Denemelerde %99,2'lik ortalama hassasiyet değerine ulaşılmıştır. Yüksek doğruluk, hızlı işlem yapısı, kararlılık ve kullanıcı dostu ara yüzün birleşimi, önerilen sistemin endüstriyel uygulamalar için umut verici bir aday olabileceğini ve sistemin gerçek dünyadaki BD üretim ortamlarında kullanıma uygunluğunu göstermektedir.

Anahtar Kelimeler

• PCB Hata Tespiti
• YOLO
• Sınıflandırma

YOLOv8-based PCB Defect Detection and Classification System

Öz

Surface inspection of Printed Circuit Boards (PCB) is one of the most crucial quality control processes due to potential serious costs of even small errors occurred during production. In this study, a YOLOv8 based system is developed for detection and classification of six common errors occurs on PCBs. In terms of accuracy, speed, and the ability to detect multiple defects simultaneously, proposed method is more suitable for use in production compared to other PCB defect detection methods. Proposed system also offers customizable defect selection for targeted inspection. Experimental results show an impressive mean average precision of 99.2%. Combination of high accuracy, fast processing speed, stability, and user-friendly interface makes it a promising candidate for industrial applications demonstrate the system's suitability for real-world PCB manufacturing environments.

Anahtar Kelimeler

• PCB Defect Detection
• YOLO
• Classification

Kaynakça

1. Markatos, N.G., & Mousavi, A., 2023. Manufacturing quality assessment in the Industry 4.0 era: A review. Total Quality Management & Business Excellence, Vol.34(13–14), pp.1655–1681. DOI:10.1080/14783363.2023.2194524.
2. Aggarwal, N., Deshwal, M., & Samant, P., 2022. A survey on automatic printed circuit board defect detection techniques. 2022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE), IEEE. DOI:10.1109/ICACITE53722.2022.9823872.
3. Bonello, D.K., Iano, Y., & Neto, U.B., 2018. A new based image subtraction algorithm for bare PCB defect detection. International Journal of Multimedia and Image Processing, Vol.8(3), pp.438–442. DOI:10.20533/ijmip.2042.4647.2018.0054.
4. Tsai, D.M., & Huang, C.K., 2018. Defect detection in electronic surfaces using template-based Fourier image reconstruction. IEEE Transactions on Components, Packaging and Manufacturing Technology, Vol.9(1), pp.163–172. DOI:10.1109/TCPMT.2018.2873744.
5. Ibrahim, Z., & Al-Attas, S.A.R., 2005. Wavelet-based printed circuit board inspection algorithm. Integrated Computer-Aided Engineering, Vol.12(2), pp.201–213. DOI:10.3233/ICA-2005-12206.
6. Zhang, Z., Wang, X., Liu, S., et al., 2018. An automatic recognition method for PCB visual defects. 2018 International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC), pp.138–142, Xi’an, China. DOI:10.1109/SDPC.2018.8664974.
7. Hassanin, I.A.A., Abd El-Samie, F.E., & El Banby, G.M., 2019. A real-time approach for automatic defect detection from PCBs based on SURF features and morphological operations. Multimedia Tools and Applications, Vol.78(24), pp.34437–34457. DOI:10.1007/s11042-019-08097-9.
8. Annaby, M., Fouda, Y., & Rushdi, M., 2019. Improved normalized cross-correlation for defect detection in printed-circuit boards. IEEE Transactions on Semiconductor Manufacturing, Vol.32(2), pp.199–211. DOI:10.1109/TSM.2019.2911062.

9. Chen, X., Wu, Y., He, X., & Ming, W., 2023. A comprehensive review of deep learning-based PCB defect detection. IEEE Access, Vol.11, pp.139017–139038. DOI:10.1109/ACCESS.2023.3339561.
10. Jiang, L., Kong, G., & Li, C., 2019. Wrapper framework for test cost-sensitive feature selection. IEEE Transactions on Systems, Man, and Cybernetics: Systems, Vol.51(3), pp.1747–1756. DOI:10.1109/TSMC.2019.2904662.
11. Jiang, L., Qiu, C., & Li, C., 2015. A novel minority cloning technique for cost-sensitive learning. International Journal of Pattern Recognition and Artificial Intelligence, Vol.29(4), pp.1551004–1551010. DOI:10.1142/S0218001415510040.
12. Zhang, H., Jiang, L., & Li, C., 2021. CS-ResNet: Cost-sensitive residual convolutional neural network for PCB cosmetic defect detection. Expert Systems with Applications, Vol.185. DOI:10.1016/j.eswa.2021.115673.
13. Hu, B., & Wang, J., 2020. Detection of PCB surface defects with improved faster-RCNN and feature pyramid network. IEEE Access, Vol.8, pp.108335–108345. DOI:10.1109/ACCESS.2020.3001349.
14. Kim, J., Ko, J., Choi, H., & Kim, H., 2021. Printed circuit board defect detection using deep learning via a skip-connected convolutional autoencoder. Sensors, Vol.21(15), p.4968. DOI:10.3390/s21154968.
15. Bhattacharya, A., & Cloutier, S.G., 2022. End-to-end deep learning framework for printed circuit board manufacturing defect classification. Scientific Reports, Vol.12(1), article 12559. DOI:10.1038/s41598-022-16302-3.
16. Chen, I.C., Hwang, R.C., & Huang, H.C., 2023. PCB defect detection based on deep learning algorithm. Processes, Vol.11(3), p.775. DOI:10.3390/pr11030775.
17. Liu, G., & Wen, H., 2021. Printed circuit board defect detection based on MobileNet-Yolo-Fast. Journal of Electronic Imaging, Vol.30(4), article 043004. DOI:10.1117/1.JEI.30.4.043004.
18. Adibhatla, V.A., Chih, H.C., Hsu, C.C., Cheng, J., Abbod, M., & Shieh, J.S., 2021. Applying deep learning to defect detection in printed circuit boards via a newest model of you-only-look-once. Mathematical Biosciences and Engineering, Vol.18(4), pp.4411–4428. DOI:10.3934/mbe.2021223.
19. Li, J., Gu, J., Huang, Z., & Wen, J., 2019. Application research of improved YOLO V3 algorithm in PCB electronic component detection. Applied Sciences, Vol.9(18), p.3750. DOI:10.3390/app9183750.
20. Mamidi, J.S.S.V., Sameer, S., & Bayana, J., 2022. A light weight version of PCB defect detection system using YOLO V4 tiny. 2022 International Mobile and Embedded Technology Conference (MECON), pp.441–445, Noida, India. DOI:10.1109/MECON53876.2022.9752361.
21. Jiang, P., et al., 2022. A review of YOLO algorithm developments. Procedia Computer Science, Vol.199, pp.1066–1073. DOI:10.1016/j.procs.2022.01.135.
22. Gillani, I.S., et al., 2022. Yolov5, Yolo-x, Yolo-r, Yolov7 performance comparison: A survey. Artificial Intelligence and Fuzzy Logic System, pp.17–28. DOI:10.5121/csit.2022.121602.
23. Hussain, M., 2023. YOLO-v1 to YOLO-v8, the rise of YOLO and its complementary nature toward digital manufacturing and industrial defect detection. Machines, Vol.11(7), p.677. DOI:10.3390/machines11070677.
24. GitHub, 2024. DeepPCB dataset. Available at: https://github.com/tangsanli5201/DeepPCB [Accessed 20 July 2024].
25. PyPI, 2024. LabelImg. Available at: https://pypi.org/project/labelImg [Accessed 20 July 2024].
26. ML CO2, 2024. Machine Learning Carbon Impact Calculator. Available at: https://mlco2.github.io/impactcomputeMachineLearningImpact [Accessed 20 July 2024].
27. Lin, T.Y., et al., 2017. Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp.2980–2988. DOI:10.1109/ICCV.2017.324.
28. Redmon, J., & Farhadi, A., 2018. YOLOv3: An incremental improvement. arXiv preprint, arXiv:1804.02767. DOI:10.48550/arXiv.1804.02767.