PCB defect detection based on YOLOV8 architecture
https://doi.org/10.21122/2309-4923-2024-2-16-24
Abstract
The paper discusses the key factors and trends in the design and production of printed circuit boards (PCB), which determine the state of the art of the automatic PCB inspection. To search for and classify defects, it is proposed to use the method of detecting defects in images based on the YOLO family of object detection models. The model was trained on a public set of images of PCB with 6 classes of defects, and the accuracy was assessed using generally accepted metrics. On the test dataset, the average accuracy according to the mAP50 metric is 0.98.
About the Authors
A. V. Inyutin
United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Belarus
Belarus
Alexander Inyutin, head of the laboratory
Minsk
M. M. Lukashevich
United Institute of Informatics Problems of the National Academy of Sciences of Belarus;
Belarusian State University
Belarus
Belarus
MarinaLukashevich,AssociateProfessor,AssociateProfessoroftheDepartmentofInformationManagementSystemsattheBelarusianStateUniversity
Minsk
References
1. What is the PCB Market Situation Now? (Update 2023). – [Online]. – Available: https://www.raypcb.com/pcb-market/ (Date of access 07.05.2024).
2. The Global Printed Circuit Board Market: Key Insights, Forecasts, & Growth Outlook. – [Online]. – Available: https://www.mktpcb.com/pcb-industry-statistics-trends-infographic/ (Date of access 07.05.2024).
3. Karpov, S. Precision Control of Printed Circuit Boards. What is this? / Sergey Karpov // Tekhnologii v elektronnykh promyshlennosti. – 2008. – № 7. – С.37–40.
4. Lehmann, David K. X-ray systems for optimizing PCB inspection: x-ray systems are not all equal, and their differences affect the type of defects that can be detected / David K. Lehmann // Circuits Assembly. – 2002. – Vol. 13, № 2. – P. 35–40.
5. Girshnick R. еt al. Rich feature hierarchies for accurate object detection and semantic segmentation [Online]. – Available: https://arxiv.org/abs/1311.2524 (Date of access 07.05.2024).
6. Girshnick R. Fast R-CNN [Online]. – Available: https://arxiv.org/abs/1504.08083 (Date of access 07.05.2024).
7. Ren S. et al. Faster R-CNN: Towards real-time object detection with region proposal networks [Online]. – Available: https://arxiv.org/abs/1506.01497 (Date of access 07.05.2024).
8. Redmon J. et al. You Only Look Once: Unified, real-time object detection [Online]. – Available: https://arxiv.org/ abs/1506.02640 (Date of access 07.05.2024).
9. Redmon J. et al. YOLO9000: Better, faster, stronger [Online]. – Available: https://arxiv.org/abs/1612.08242 (Date of access 07.05.2024).
10. Redmon J. YOLOv3: An incremental improvement [Online]. – Available: https://arxiv.org/abs/1804.02767 (Date of access 07.05.2024).
11. YOLOv8: A New State-of-the-Art Computer Vision Model. – [Online]. – Available: https://yolov8.com/ (Date of access 07.05.2024).
12. A PCB defect dataset. – [Online]. – Available: https://github.com/tangsanli5201/DeepPCB (Date of access 07.05.2024).
13. DeepPCB Image Dataset. – [Online]. – Available: https://universe.roboflow.com/tack-hwa-wong-zak5u/deeppcb4dhir/dataset/5 (Date of access 07.05.2024).
14. Configuration Ultralytics YOLOv8 Docs. – [Online]. – Available: https://docs.ultralytics.com/usage/cfg/#modes (Date of access 07.05.2024).
Review
For citations:
Inyutin A.V., Lukashevich M.M. PCB defect detection based on YOLOV8 architecture. «System analysis and applied information science». 2024;(2):16-24. (In Russ.) https://doi.org/10.21122/2309-4923-2024-2-16-24
Views:
301
Email this article 