PERBAIKAN LOKAL KANDIDAT CACAT TANPA PEMBELAJARAN UNTUK INSPEKSI PCB BERBASIS TEMPLATE DENGAN KLASIFIKASI ROI SELEKTIF
-
Bernadectus Yudi Dwiandiyanta(1*)
Universitas Atma Jaya Yogyakarta
(*) Corresponding Author
Abstract
Printed circuit board (PCB) defect inspection is difficult because many defects are small and visually subtle. This study proposes a template-based inspection method combining image differencing, learning-free local candidate refinement, and selective ROI classification. Initial candidates are obtained from aligned test and template images, then locally refined using grayscale residual information without adding new candidates. Refined ROIs are represented by test, template, and residual patches and classified using MobileNetV3-Small. On DeepPCB, candidate recall at IoU 0.20 increases from 0.3389 to 0.9586 and at IoU 0.30 from 0.1153 to 0.6344. ROI classification achieves 0.9874 accuracy and 0.9855 macro F1. With a rejection threshold of 0.90, coverage remains 0.9762 with macro F1 of 0.9927. The method improves coarse defect localization, although strict bounding-box localization remains limited.
Downloads
Download data is not yet available.
References
[1] H. Guo, H. Zhao, Y. Zhao, and W. Liu, “PCB defect detection algorithm based on deep learning,” Optik (Stuttg)., vol. 315, pp. 1–13, 2024, doi: 10.1016/j.ijleo.2024.172036.
[2] P. Chen and F. Xie, “A Machine Learning Approach for Automated Detection of Critical PCB Flaws in Optical Sensing Systems,” Photonics, vol. 10, no. 9, 2023, doi: 10.3390/photonics10090984.
[3] J.-H. Park, Y.-S. Kim, H. Seo, and Y.-J. Cho, “Analysis of Training Deep Learning Models for PCB Defect Detection,” Sensors, vol. 23, no. 5, p. 2766, Mar. 2023, doi: 10.3390/s23052766.
[4] J. Onshaunjit and J. Srinonchat, “Algorithmic scheme for concurrent detection and classification of printed circuit board defects,” Comput. Mater. Contin., vol. 71, no. 1, pp. 355–367, 2022, doi: 10.32604/cmc.2022.017698.
[5] X. Chen, Y. Wu, X. He, and W. Ming, “A Comprehensive Review of Deep Learning-Based PCB Defect Detection,” IEEE Access, vol. 11, no. December, pp. 139017–139038, 2023, doi: 10.1109/ACCESS.2023.3339561.
[6] T. T. A. Pham, D. K. T. Thoi, H. Choi, and S. Park, “Defect Detection in Printed Circuit Boards Using Semi-Supervised Learning,” Sensors, vol. 23, no. 6, 2023, doi: 10.3390/s23063246.
[7] B. Feng and J. Cai, “PCB Defect Detection via Local Detail and Global Dependency Information,” Sensors, vol. 23, no. 18, pp. 1–18, 2023, doi: 10.3390/s23187755.
[8] Y. Jiang, M. Cai, and D. Zhang, “Lightweight Network DCR-YOLO for Surface Defect Detection on Printed Circuit Boards,” Sensors, vol. 23, no. 17, 2023, doi: 10.3390/s23177310.
[9] B. Du, F. Wan, G. Lei, L. Xu, C. Xu, and Y. Xiong, “YOLO-MBBi: PCB Surface Defect Detection Method Based on Enhanced YOLOv5,” Electron., vol. 12, no. 13, pp. 1–24, 2023, doi: 10.3390/electronics12132821.
[10] G. Xiao, S. Hou, and H. Zhou, “PCB defect detection algorithm based on CDI-YOLO,” Sci. Rep., vol. 14, no. 1, pp. 1–12, 2024, doi: 10.1038/s41598-024-57491-3.
[11] T. Zhang, J. Zhang, P. Pan, and X. Zhang, “YOLO-RRL: A Lightweight Algorithm for PCB Surface Defect Detection,” Appl. Sci., vol. 14, no. 17, 2024, doi: 10.3390/app14177460.
[12] Y. Gao, Z. Li, Y. Wang, and S. Zhu, “A Novel YOLOv5_ES based on lightweight small object detection head for PCB surface defect detection,” Sci. Rep., vol. 14, no. 1, pp. 1–12, 2024, doi: 10.1038/s41598-024-74368-7.
[13] T. Kiobya, J. Zhou, B. Maiseli, and M. Khan, “Attentive context and semantic enhancement mechanism for printed circuit board defect detection with two-stage and multi-stage object detectors,” Sci. Rep., vol. 14, no. 1, pp. 1–18, 2024, doi: 10.1038/s41598-024-69207-8.
[14] W. Chen, S. Meng, and X. Wang, “Local and Global Context-Enhanced Lightweight CenterNet for PCB Surface Defect Detection,” Sensors, vol. 24, no. 14, p. 4729, Jul. 2024, doi: 10.3390/s24144729.
[15] J. Peng, W. Fan, S. Lan, and D. Wang, “MDD-DETR: Lightweight Detection Algorithm for Printed Circuit Board Minor Defects,” Electron., vol. 13, no. 22, 2024, doi: 10.3390/electronics13224453.
[16] D. Lang and Z. Lv, “SEPDNet: simple and effective PCB surface defect detection method,” Sci. Rep., vol. 15, no. 1, pp. 1–15, 2025, doi: 10.1038/s41598-024-84859-2.
[17] Z. F. Elsharkawy, “Enhanced YOLOv11 framework for high precision defect detection in printed circuit boards,” Sci. Rep., vol. 15, no. 1, pp. 1–17, 2025, doi: 10.1038/s41598-025-27415-w.
[18] Y. Wang et al., “Enhanced PCB defect detection via HSA-RTDETR on RT-DETR,” Sci. Rep., vol. 15, no. 1, pp. 1–19, 2025, doi: 10.1038/s41598-025-11394-z.
[19] L. Zhu and R. Zhao, “A novel PCB surface defect detection method based on separated global context attention to guide residual context aggregation,” Sci. Rep., vol. 15, no. 1, pp. 1–18, 2025, doi: 10.1038/s41598-024-84961-5.
[20] W. Chen, X. Ke, and S. Meng, “Small defect detection in printed circuit boards based on the multiscale edge strengthening and an improved YOLOv10,” Sci. Rep., vol. 15, no. 1, pp. 1–15, 2025, doi: 10.1038/s41598-025-20387-x.
[21] L. Bai and W. H. Xu, “Improved printed circuit board defect detection scheme,” Sci. Rep., vol. 15, no. 1, pp. 1–13, 2025, doi: 10.1038/s41598-025-85245-2.
[22] Y. Yan, G. Chen, Y. Gu, and X. Peng, “Hybrid CNN-transformer with adaptive multi-scale fusion for tiny PCB defect identification,” Complex Intell. Syst., 2026, doi: 10.1007/s40747-026-02326-z.
[23] Z. Shen, Y. Yao, L. Liu, Y. Cao, and L. Lu, “Defect detection method of printed circuit boards based on EDF-YOLOv10,” PLoS One, vol. 21, no. 3 March, pp. 1–19, 2026, doi: 10.1371/journal.pone.0343130.
[2] P. Chen and F. Xie, “A Machine Learning Approach for Automated Detection of Critical PCB Flaws in Optical Sensing Systems,” Photonics, vol. 10, no. 9, 2023, doi: 10.3390/photonics10090984.
[3] J.-H. Park, Y.-S. Kim, H. Seo, and Y.-J. Cho, “Analysis of Training Deep Learning Models for PCB Defect Detection,” Sensors, vol. 23, no. 5, p. 2766, Mar. 2023, doi: 10.3390/s23052766.
[4] J. Onshaunjit and J. Srinonchat, “Algorithmic scheme for concurrent detection and classification of printed circuit board defects,” Comput. Mater. Contin., vol. 71, no. 1, pp. 355–367, 2022, doi: 10.32604/cmc.2022.017698.
[5] X. Chen, Y. Wu, X. He, and W. Ming, “A Comprehensive Review of Deep Learning-Based PCB Defect Detection,” IEEE Access, vol. 11, no. December, pp. 139017–139038, 2023, doi: 10.1109/ACCESS.2023.3339561.
[6] T. T. A. Pham, D. K. T. Thoi, H. Choi, and S. Park, “Defect Detection in Printed Circuit Boards Using Semi-Supervised Learning,” Sensors, vol. 23, no. 6, 2023, doi: 10.3390/s23063246.
[7] B. Feng and J. Cai, “PCB Defect Detection via Local Detail and Global Dependency Information,” Sensors, vol. 23, no. 18, pp. 1–18, 2023, doi: 10.3390/s23187755.
[8] Y. Jiang, M. Cai, and D. Zhang, “Lightweight Network DCR-YOLO for Surface Defect Detection on Printed Circuit Boards,” Sensors, vol. 23, no. 17, 2023, doi: 10.3390/s23177310.
[9] B. Du, F. Wan, G. Lei, L. Xu, C. Xu, and Y. Xiong, “YOLO-MBBi: PCB Surface Defect Detection Method Based on Enhanced YOLOv5,” Electron., vol. 12, no. 13, pp. 1–24, 2023, doi: 10.3390/electronics12132821.
[10] G. Xiao, S. Hou, and H. Zhou, “PCB defect detection algorithm based on CDI-YOLO,” Sci. Rep., vol. 14, no. 1, pp. 1–12, 2024, doi: 10.1038/s41598-024-57491-3.
[11] T. Zhang, J. Zhang, P. Pan, and X. Zhang, “YOLO-RRL: A Lightweight Algorithm for PCB Surface Defect Detection,” Appl. Sci., vol. 14, no. 17, 2024, doi: 10.3390/app14177460.
[12] Y. Gao, Z. Li, Y. Wang, and S. Zhu, “A Novel YOLOv5_ES based on lightweight small object detection head for PCB surface defect detection,” Sci. Rep., vol. 14, no. 1, pp. 1–12, 2024, doi: 10.1038/s41598-024-74368-7.
[13] T. Kiobya, J. Zhou, B. Maiseli, and M. Khan, “Attentive context and semantic enhancement mechanism for printed circuit board defect detection with two-stage and multi-stage object detectors,” Sci. Rep., vol. 14, no. 1, pp. 1–18, 2024, doi: 10.1038/s41598-024-69207-8.
[14] W. Chen, S. Meng, and X. Wang, “Local and Global Context-Enhanced Lightweight CenterNet for PCB Surface Defect Detection,” Sensors, vol. 24, no. 14, p. 4729, Jul. 2024, doi: 10.3390/s24144729.
[15] J. Peng, W. Fan, S. Lan, and D. Wang, “MDD-DETR: Lightweight Detection Algorithm for Printed Circuit Board Minor Defects,” Electron., vol. 13, no. 22, 2024, doi: 10.3390/electronics13224453.
[16] D. Lang and Z. Lv, “SEPDNet: simple and effective PCB surface defect detection method,” Sci. Rep., vol. 15, no. 1, pp. 1–15, 2025, doi: 10.1038/s41598-024-84859-2.
[17] Z. F. Elsharkawy, “Enhanced YOLOv11 framework for high precision defect detection in printed circuit boards,” Sci. Rep., vol. 15, no. 1, pp. 1–17, 2025, doi: 10.1038/s41598-025-27415-w.
[18] Y. Wang et al., “Enhanced PCB defect detection via HSA-RTDETR on RT-DETR,” Sci. Rep., vol. 15, no. 1, pp. 1–19, 2025, doi: 10.1038/s41598-025-11394-z.
[19] L. Zhu and R. Zhao, “A novel PCB surface defect detection method based on separated global context attention to guide residual context aggregation,” Sci. Rep., vol. 15, no. 1, pp. 1–18, 2025, doi: 10.1038/s41598-024-84961-5.
[20] W. Chen, X. Ke, and S. Meng, “Small defect detection in printed circuit boards based on the multiscale edge strengthening and an improved YOLOv10,” Sci. Rep., vol. 15, no. 1, pp. 1–15, 2025, doi: 10.1038/s41598-025-20387-x.
[21] L. Bai and W. H. Xu, “Improved printed circuit board defect detection scheme,” Sci. Rep., vol. 15, no. 1, pp. 1–13, 2025, doi: 10.1038/s41598-025-85245-2.
[22] Y. Yan, G. Chen, Y. Gu, and X. Peng, “Hybrid CNN-transformer with adaptive multi-scale fusion for tiny PCB defect identification,” Complex Intell. Syst., 2026, doi: 10.1007/s40747-026-02326-z.
[23] Z. Shen, Y. Yao, L. Liu, Y. Cao, and L. Lu, “Defect detection method of printed circuit boards based on EDF-YOLOv10,” PLoS One, vol. 21, no. 3 March, pp. 1–19, 2026, doi: 10.1371/journal.pone.0343130.
Published
2026-10-14
How to Cite
[1]
B. Y. Dwiandiyanta, “PERBAIKAN LOKAL KANDIDAT CACAT TANPA PEMBELAJARAN UNTUK INSPEKSI PCB BERBASIS TEMPLATE DENGAN KLASIFIKASI ROI SELEKTIF”, JME, Oct. 2026.
Section
Articles
This work is licensed under CC BY-SA 4.0
