YOLO Destekli Görüntü İşleme ile Otomotiv Şasesinde Simülasyon Tabanlı Punta Kaynak Muayenesi

Öz

Bu çalışma, otomotiv endüstrisinde yaygın olarak kullanılan Direnç Nokta Kaynağı (RSW) kalite kontrol süreçlerini iyileştirmeyi amaçlayan simülasyon tabanlı bir test platformunu incelemektedir. Fiziksel prototiplere olan ihtiyacı azaltmak için sanal bir test ortamı geliştirilmiştir. Platform, simülasyon ortamı içerisinde RoboDK kütüphanesinden elde edilen bir araç şasesi üzerine ESP32-CAM tabanlı sanal kameralar yerleştirilerek oluşturulmuştur. Kaggle’dan elde edilen yaklaşık 1.000 gerçek RSW görüntüsünden oluşan veri seti Roboflow kullanılarak etiketlenmiş ve YOLO (You Only Look Once) mimarisiyle uyumlu formata dönüştürülmüştür. Görüntü işleme ve nesne tanıma aşamasında YOLOv3-s ve YOLOv5-m modelleri kullanılmıştır. Modellerin sınıflandırma performansları F1 skoru, kesinlik, duyarlılık, ortalama doğruluk (mAP) ve güven skoru gibi metriklerle değerlendirilmiştir. Her iki model de düşük donanım gereksinimleri ile çalışabilmekle birlikte, YOLOv5-m genel olarak daha üstün performans sergilemiştir. Özellikle kritik kaynak kusurları olarak sınıflandırılan Sınıf 2 (patlama kaynağı) tespitinde YOLOv5-m modeli daha yüksek güven skorları elde etmiştir. Bu simülasyon tabanlı yöntem, RSW kalite kontrolünü daha hızlı, daha ekonomik ve daha güvenilir hale getirmiştir. Sonuç olarak, robotik kaynak sistemleri üretim hattına entegre edilmeden önce doğruluk ve güvenlik açısından sanal ortamda kapsamlı şekilde test edilebilmiştir.

Anahtar Kelimeler

• otomotiv endüstrisi
• derin öğrenme
• direnç nokta kaynağı
• görüntü işleme
• RoboDK
• YOLO

Simulation-based spot welding inspection on automotive chassis using YOLO-powered image processing

Öz

This study examines a simulation-based testing platform designed to enhance the quality control processes of Resistance Spot Welding (RSW), a technology widely used in the automotive industry. A virtual testing environment was developed to eliminate the need for physical prototypes. The platform was assembled by placing ESP32-CAM-based virtual cameras on a vehicle chassis obtained from the RoboDK library within the simulation environment. A dataset of approximately 1,000 real RSW images from Kaggle was labeled using Roboflow and converted into a format compatible with YOLO(You Only Look Once) architecture. During image processing and object recognition, YOLOv3-s and YOLOv5-m models were utilized. The models’ classification performance was evaluated using metrics such as F1 score, precision, recall, mean average precision (mAP), and Confidence Score (CS). Both models required low hardware requirements; however, YOLOv5-m displayed overall superior performance. Notably, the YOLOv5-m model achieved higher confidence scores in detecting critical welding defects classified as Class 2 (explosion weld); an approximate increase of 8–9% was observed in experimental results, reaching a CS of around 0.58. In addition, the F1 score for Class 2 (explosion weld) improved by approximately 5–6%, reaching a value of around 0.85. This simulation-based method has made RSW quality control faster, more cost-effective, and reliable. Consequently, robotic welding systems can be thoroughly tested for accuracy and safety in a virtual environment before being integrated into the production line.

Anahtar Kelimeler

• Automotive Industry
• Deep Learning
• Image Processing
• Resistance Spot Welding
• RoboDK
• YOLO

Kaynakça

1. Capezza, C., Centofanti, F., Lepore, A., & Palumbo, B., Functional clustering methods for resistance spot welding process data in the automotive industry. Applied Stochastic Models in Business and Industry, 37 (5), 908-925, 2021. https://doi.org/10.1002/asmb.2648
2. Li, D., Yang, P., & Zou, Y., Optimizing insulator defect detection with improved DETR models. Mathematics, 12 (10), 1507 2024. https://doi.org/10.3390/math12101507
3. Dai, W., Li, D., Zheng, Y., Wang, D., Tang, D., Wang, H., & Peng, Y. Online quality inspection of resistance spot welding for automotive production lines. Journal of Manufacturing Systems, 63, 354-369, 2022. https://doi.org/10.1016/j.jmsy.2022.04.008
4. Mathiszik, C., Köberlin, D., Heilmann, S., Zschetzsche, J., & Füssel, U., General approach for inline electrode wear monitoring at resistance spot welding. Processes, 9 (4), 685, 2021. https://doi.org/10.3390/pr9040685
5. Liu, W., Hu, J., & Qi, J., Resistance Spot Welding Defect Detection Based on Visual Inspection: Improved Faster R-CNN Model. Machines, 13 (1), 33, 2025. https://doi.org/10.3390/machines13010033
6. Wang, X. J., Zhou, J. H., Yan, H. C., & Pang, C. K. Quality monitoring of spot welding with advanced signal processing and data-driven techniques. Transactions of the Institute of Measurement and Control, 40 (7), 2291-2302, 2018. https://doi.org/10.1177/0142331217700703
7. Yu, X., Sun, X., & Ou, L., Graphics-based modular digital twin software framework for production lines. Computers & Industrial Engineering, 193, 110308, 2024. https://doi.org/10.1016/j.cie.2024.110308
8. Wang, Z., Zhang, M., & Xu, Y., Development of a robotic arm control platform for ultrasonic testing inspection in remanufacturing. In 2022 27th International Conference on Automation and Computing (ICAC), (pp. 1-6). IEEE, 2022. https://doi.org/10.1109/ICAC55051.2022.9911174

9. Gheorghe, C., Duguleana, M., Boboc, R. G., & Postelnicu, C. C., Analyzing Real-Time Object Detection with YOLO Algorithm in Automotive Applications: A Review. CMES-Computer Modeling in Engineering & Sciences, 141 (3), 2024. https://doi.org/10.32604/cmes.2024.054735
10. Singh, A., Kalaichelvi, V., DSouza, A., & Karthikeyan, R., GAN-Based image dehazing for intelligent weld shape classification and tracing using deep learning. Applied Sciences, 12 (14), 6860, 2022. https://doi.org/10.3390/app12146860
11. Lang, X., Ren, Z., Wan, D., Zhang, Y., & Shu, S. (MR-YOLO: An improved YOLOv5 network for detecting magnetic ring surface defects. Sensors, 22 (24), 9897, 2024. https://doi.org/10.3390/s22249897
12. Elhattab, K., Abouelmehdi, K., & Elatar, S, New model to monitor plant growth remotely using esp32-cam and mobile application. In 2023 10th International Conference on Wireless Networks and Mobile Communications (WINCOM) IEEE.,(pp. 1-6), 2023. https://doi.org/10.1109/WINCOM59760.2023.10322939
13. Ragab, M. G., Abdulkader, S. J., Muneer, A., Alqushaibi, A., Sumiea, E. H., Qureshi, R., ... & Alhussian, H., A comprehensive systematic review of YOLO for medical object detection (2018 to 2023). IEEE Access., 2024. https://doi.org/10.1109/ACCESS.2024.3386826
14. Ramchandani, M., Sahu, S. P., & Dewangan, D. K., A comparative study in pedestrian detection for autonomous driving systems. In 2022 OPJU International Technology Conference on Emerging Technologies for Sustainable Development (OTCON) IEEE. (pp. 1-6), 2023. https://doi.org/10.1109/OTCON56053.2023.10113992
15. Domínguez, L., Rivas-Araiza, E. A., Jáuregui-Correa, J. C., González-Córdoba, J. L., Pedraza-Ortega, J. C., & Takács, A, Resistance spot welding insights: A dataset integrating process parameters, infrared, and surface imaging. Data in Brief, 59, 111373, 2025. https://doi.org/10.1016/j.dib.2025.111373
16. Karaman, A., Karaboga, D., Pacal, I., Akay, B., Basturk, A., Nalbantoglu, U.,& Sahin, O., Hyper-parameter optimization of deep learning architectures using artificial bee colony (ABC) algorithm for high performance real-time automatic colorectal cancer (CRC) polyp detection. Applied Intelligence, 53 (12), 15603-15620, 2023. https://doi.org/10.1007/s10489-022-04299-1
17. Gullino, A., Matteis, P., & D’Aiuto, F., Review of aluminum-to-steel welding technologies for car-body applications. Metals, 9 (3), 315, 2019. https://doi.org/10.3390/met9030315
18. Guo, K., Sui, L., Qiu, J., Yu, J., Wang, J., Yao, S., ... & Yang, H., Angel-eye: A complete design flow for mapping CNN onto embedded FPGA. IEEE transactions on computer-aided design of integrated circuits and systems, 37 (1), 35-47, 2017. https://doi.org/10.1109/TCAD.2017.2705069
19. Kulikov, A. A., Sidorova, A. V., & Balanovskii, A. E. Programming industrial robots for wire arc additive manufacturing. In International Conference on Industrial Engineering Cham: Springer International Publishing, (pp. 566-576), 2021. https://doi.org/10.1007/978-3-030-54817-9_66
20. Zhou, S., Ao, S., Yang, Z., & Liu, H., Surface defect detection of steel plate based on SKS-YOLO. IEEE Access, 2024. https://doi.org/10.1109/ACCESS.2024.3422244
21. Singh, A., Raj, K., Kumar, T., Verma, S., & Roy, A. M., Deep learning-based cost-effective and responsive robot for autism treatment. Drones, 7 (2), 81, 2023. https://doi.org/10.3390/drones7020081
22. Sozzi, M., Cantalamessa, S., Cogato, A., Kayad, A., & Marinello, F., Automatic bunch detection in white grape varieties using YOLOv3, YOLOv4, and YOLOv5 deep learning algorithms. Agronomy, 12 (2), 319, 2022. https://doi.org/10.3390/agronomy12020319
23. Wu, J., Shen, T., Wang, Q., Tao, Z., Zeng, K., & Song, J. Local adaptive illumination-driven input-level fusion for infrared and visible object detection. Remote Sensing, 15 (3), 660, 2023. https://doi.org/10.3390/rs15030660
24. Swain, S., & Tripathy, A. K., Automatic detection of potholes using VGG-16 pre-trained network and Convolutional Neural Network. Heliyon, 10 (10), 2024. https://doi.org/10.1016/j.heliyon.2024.e30957
25. Mohammadrezaei, E., Ghasemi, S., Dongre, P., Gračanin, D., & Zhang, H. Systematic review of extended reality for smart built environments lighting design simulations. IEEE Access, 2024; 12, 17058-17089, 2024. https://doi.org/10.1109/ACCESS.2024.3359167
26. Kshirsagar, V., Bhalerao, R. H., & Chaturvedi, M. Modified yolo module for efficient object tracking in a video. IEEE Latin America Transactions, 21 (3), 389-398, 2023. https://doi.org/10.1109/TLA.2023.10068842
27. Glučina, M., Anđelić, N., Lorencin, I., & Car, Z., Detection and classification of printed circuit boards using YOLO algorithm. Electronics, 12 (3), 667, 2023. https://doi.org/10.3390/electronics12030667
28. Long, X., Deng, K., Wang, G., Zhang, Y., Dang, Q., Gao, Y., & Wen, S., PP-YOLO: An effective and efficient implementation of object detector arXiv preprint arXiv:2007.12099, 2020. https://doi.org/10.48550/arXiv.2007.12099
29. Wu, S., Li, X., & Wang, X., IoU-aware single-stage object detector for accurate localization. Image and Vision Computing, 97, 103911, 2020. https://doi.org/10.1016/j.imavis.2020.103911
30. Sotres, J., Boyd, H., & Gonzalez-Martinez, J. F., Enabling autonomous scanning probe microscopy imaging of single molecules with deep learning. Nanoscale, 13 (20), 9193-9203, 2021. https://doi.org/10.1039/D1NR01109J