SavedAdd your journal

Journals / Muş Alparslan Üniversitesi Fen Bilimleri Dergisi / 2026 / Cilt: 14 Sayı: 1

Comparative Analysis of YOLOv10 and YOLOv11 Architectures for Printed Circuit Board Defect Detection

Baskı Devre Kartı Kusur Tespiti için YOLOv10 ve YOLOv11 Mimarilerinin Karşılaştırmalı Analizi

Züleyha Kezer(FIRAT ÜNİVERSİTESİ)0009-0009-1836-8077
Feyza Altunbey Özbay(FIRAT ÜNİVERSİTESİ)0000-0003-0629-6888
Pages
162–172
Publication Date
2026-06-27
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results showed that the YOLOv10-Medium model trained with SGD achieved the highest accuracy with a value of 70.30 mAP@50:95.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Detecting defects in Printed Circuit Boards (PCBs) is critical for production efficiency and quality control.
📊
Key Findings & Data:This study compares the deep learning-based object detection architectures YOLOv10 and YOLOv11 for PCB defect detection.
💡
Conclusion & Impact:The results showed that the YOLOv10-Medium model trained with SGD achieved the highest accuracy with a value of 70.30 mAP@50:95.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Baskı Devre Kartı · Derin Öğrenme · Kusur Tespiti

Abstract

Detecting defects in Printed Circuit Boards (PCBs) is critical for production efficiency and quality control. This study compares the deep learning-based object detection architectures YOLOv10 and YOLOv11 for PCB defect detection. Nano, Small, and Medium variations of both architectures were used in the analyses, and their performance was evaluated using SGD and Adamax optimization algorithms. Experiments were conducted on a PKU-Market-PCB dataset containing six defect categories and 10,668 images. The results showed that the YOLOv10-Medium model trained with SGD achieved the highest accuracy with a value of 70.30 mAP@50:95. However, this model has a higher computational cost. When accuracy and computational efficiency are considered together, it was determined that Small-scale models offer a more balanced solution for industrial applications. Furthermore, it was found that YOLOv10 provided an average of 1.53% higher performance than YOLOv11, and SGD provided an average of 5.33% higher performance than Adamax. The aim of these experiments is to determine the appropriate model-optimization combination for PCB defect detection.

Özet

Baskı Devre Kartı (Printed Circuit Board-PCB) kusurlarının tespiti, üretim verimliliği ve kalite kontrol açısından kritik öneme sahiptir. Bu çalışmada, PCB kusur tespiti için derin öğrenme tabanlı nesne algılama mimarileri YOLOv10 ve YOLOv11 karşılaştırılmıştır. Analizlerde her iki mimarinin Nano, Small ve Medium varyasyonları kullanılmış, performansları SGD ve Adamax optimizasyon algoritmalarıyla değerlendirilmiştir. Deneyler, altı kusur kategorisi içeren ve 10.668 görüntüden oluşan PKU-Market-PCB veri kümesi üzerinde gerçekleştirilmiştir. Sonuçlar, SGD ile eğitilen YOLOv10-Medium modelinin %70.30 mAP@50:95 değeriyle en yüksek doğruluğu sağladığını göstermiştir. Ancak bu model daha yüksek hesaplama maliyetine sahiptir. Doğruluk ve hesaplama verimliliği birlikte değerlendirildiğinde, Small ölçekli modellerin endüstriyel uygulamalar için daha dengeli bir çözüm sunduğu belirlenmiştir. Ayrıca YOLOv10’un YOLOv11’e göre ortalama %1.53, SGD’nin ise Adamax’a göre ortalama %5.33 daha yüksek performans sağladığı tespit edilmiştir. Gerçekleştirilen deneyler doğrultusunda, PCB kusur tespiti için uygun model-optimizasyon kombinasyonunun belirlenmesini amaçlamaktadır.

🔖

Cite This Article

Kezer, Z., Özbay, F. A. (2026). Comparative Analysis of YOLOv10 and YOLOv11 Architectures for Printed Circuit Board Defect Detection. Muş Alparslan Üniversitesi Fen Bilimleri Dergisi, Cilt: 14 Sayı: 1, 162–172. https://doi.org/10.18586/msufbd.1908942

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (23)

  1. Saberironaghi A., Ren J., El-Gindy M., Defect detection methods for industrial products using deep learning techniques: A review, Algorithms, 16(2), 2023, p.95.
  2. Zheng X., Zheng S., Kong Y. ve Chen J. Recent advances in surface defect inspection of industrial products using deep learning techniques, The International Journal of Advanced Manufacturing Technology, 113(1), p. 35-58, 2021.
  3. Chaudhary V., Dave I. R., Upla K. P., Automatic visual inspection of printed circuit board for defect detection and classification, 2017 International Conference on Wireless Communications, 2017.
  4. Khanam R., Hussain M., Hill R., Allen P., A comprehensive review of convolutional neural networks for defect detection in industrial applications, IEEE Access, 2024.
  5. Qi S., Yang J., Zhong Z., A review on industrial surface defect detection based on deep learning technology, 2020 3rd International Conference on Machine Learning and Machine Intelligence, p.24-30,2020.
  6. Chen X., Wu Y., He X., Ming W., A comprehensive review of deep learning-based PCB defect detection, IEEE Access, 11, p.139017-139038 2023.
  7. Du B., Wan F., Lei G., Xu L., Xu C., Xiong Y., YOLO-MBBi: PCB surface defect detection method based on enhanced YOLOv5, Electronics, 12(13), p.2821, 2023.
  8. Xiao G., Hou S., Zhou H., PCB defect detection algorithm based on CDI-YOLO, Scientific Reports, 14(1), p.7351. 2024.
  9. Zhou W., Li C., Ye Z., He Q., Ming Z., Chen J., Xiao Z., An efficient tiny defect detection method for PCB with improved YOLO through a compression training strategy, IEEE Transactions on Instrumentation and Measurement, 73, p.1-14, 2024.
  10. Zhang H., Li Y., Kayes D. M. S., Song Z., Wang Y., Research on PCB defect detection algorithm based on LPCB-YOLO, Frontiers in Physics, 12, p.1472584, 2025.
  11. Yuan M., Zhou Y., Ren X., Zhi H., Zhang J., Chen H., YOLO-HMC: An improved method for PCB surface defect detection, IEEE Transactions on Instrumentation and Measurement, 73, p.1-11, 2024.
  12. Budak İ., Bal S., Korkmaz H., PCB üretiminde çok sınıflı kusur tespiti için yolo tabanlı derin öğrenme modeli, Afyon Kocatepe Üniversitesi Fen Ve Mühendislik Bilimleri Dergisi, 25(4), p.816-826, 2025.
  13. Ding R., Dai L., Li G., Liu, H., TDD‐net: a tiny defect detection network for printed circuit boards, CAAI Transactions on Intelligence Technology, 4(2), p.110-116, 2019
  14. Wang J., Ma L., Li Z., Cao Y., Zhang H., SSHP-YOLO: A high precision printed circuit board (PCB) defect detection algorithm with a small sample, Electronics, 14(2), p. 217, 2025.
  15. Yang H., Dong J., Wang C., Lian Z., Chang, H., PCES-YOLO: High-Precision PCB Detection via Pre-Convolution Receptive Field Enhancement and Geometry-Perception Feature Fusion, Applied Sciences, 15(13), p. 7588, 2025.
  16. Elter N., PCB Defect Dataset, https://www.kaggle.com/datasets/norbertelter/pcb-defect-dataset, Erişim Tarihi: 10.02.2026.
  17. Allgaier J., Pryss R., Cross-Validation Visualized: A Narrative Guide to Advanced Methods, Machine Learning and Knowledge Extraction, 6(2), p.1378-1388, 2024.
  18. Wang A., Chen H., Liu L., Chen K., Lin Z., Han J., YOLOv10: Real-time end-to-end object detection, Advances in Neural Information Processing Systems, 37, 2024.
  19. He L. H., Zhou Y. Z., Liu L., Cao W., Ma J. H., Research on object detection and recognition in remote sensing images based on YOLOv11, Scientific Reports, 15(1), p.14032, 2025.
  20. Mao M., Hong M., YOLO object detection for real-time defect inspection: A review of YOLOv1 to YOLOv11, Sensors, 25(7), p.2270, 2025.
  21. Diwan T., Anirudh G., Tembhurne J. V., Object detection using YOLO: Challenges, architectural successors, datasets and applications, Multimedia Tools and Applications, 82(6), p.9243-9275, 2023.
  22. Liu C., Tao Y., Liang J., Li K., Chen Y., Object detection based on YOLO network, 2018 IEEE 4th Information Technology and Mechatronics Engineering Conference, p.799-803, 2018.
  23. Tan F. G., Yüksel A. S., Aydemir E., Ersoy M., Derin öğrenme teknikleri ile nesne tespiti ve takibi üzerine bir inceleme, Avrupa Bilim ve Teknoloji Dergisi, 25, p.159-171, 2021.