Viksa OMK, Vyksa, Russia

M. N. Shamshin, Head of Laboratory, Engineering and Technology Center, e-mail: Shamshin_MN@omk.ru
I. V. Mikheev, Chief Specialist for Product Implementation, Engineering and Technology Center, e-mail: miheev_iv@omk.ru
A. V. Rybakov, Cand. Phys.-Math., Chief Specialist for Computer Vision, Engineering and Technology Center, e-mail: rybakov_av1@omk.ru

Astrakhan State University named after V. N. Tatishchev, Astrakhan, Russia
R. Yu. Demina, Cand. Eng., Associate Prof., e-mail: raisa.demina.91@mail.ru

Technologically complex and potentially dangerous industries, which include ferrous metallurgy, have a number of tasks that can potentially be solved using artificial intelligence methods. One of these tasks is the problem of timely detection of surfacedefects in large-diameter pipes. A number of requirements are put forward for a computer vision model designed to facilitate the work of a flaw detector: timely detection, correct classification and a low number of false positives. The paper presents a methodology for developing a system for detecting and classifying surface defects based on computer vision for large-diameter pipes. As a result of the analysis of publications on the topic of defect recognition using computer vision methods and based on data obtained in real production conditions at the site of the final inspection of pipes of the finishing section of the Vyksa OMK plant, general approaches to software design, metrics for assessing detection and classification models and the sequence of personnel actions are formulated. The paper presents a comparative analysis of various approaches to building computer vision models. A scheme for calculating the resolution and number of cameras for efficient use of technical capacities is proposed. The software architecture together with the proposed approach to marking a training set allows for more productive and efficient collection of the training set. The developed methods and approaches are currently being tested and contribute to the accumulation of data for subsequent evaluation of their effectiveness.

1. Zinyagin A. G., Muntin A. V., Tynchenko V. S., Zhikharev P. I. et al. Recurrent neural network (RNN)-based approach to predict mean flow stress in industrial rolling. Metals. 2024. Vol. 14. 1329. DOI: 10.3390/met14121329
2. Muravyov S. V., Pogadaeva E. Yu. Computer-aided recognition of defects in welded joints during visual inspections based on geometric attributes. Russ. J. Nondestruct. Test. 2020. Vol. 56, Iss. 3. pp. 259–267. DOI: 10.1134/S1061830920030055
3. Sun J., Li C., Wu X. J., Palade V., Fang W. An effective method of weld defect detection and classification based on machine vision. IEEE Trans. Industr. Inform. 2019. Vol. 15, Iss. 12. pp. 6322–6333. DOI: 10.1109/TII.2019.2896357
4. Fan X., Gao X., Liu G., Ma N., Zhang Y. Research and prospect of welding monitoring technology based on machine vision. Int. J. Adv. Manuf. Technol. 2021. Vol. 115. pp. 3365–3391. DOI: 10.1007/s00170-021-07398-4
5. Li L., Xiao L., Liao H., Liu S., Ye B. Welding quality monitoring of high frequency straight seam pipe based on image feature. J. Mater. Process Technol. 2017. Vol. 246. pp. 285–290. DOI: 10.1016/j.jmatprotec.2017.03.031
6. Hou W., Zhang D., Wei Y., Guo J., Zhang X. Review on computer aided weld defect detection from radiography images. Appl. Sci. 2020. Vol. 10, Iss. 5. 1878. DOI: 10.3390/app10051878
7. Boaretto N., Centeno T. M. Automated detection of welding defects in pipelines from radiographic images DWDI. NDT & E International. 2017. Vol. 86. pp. 7–13. DOI: 10.1016/j.ndteint.2016.11.003
8. Valavanis I., Kosmopoulos D. Multiclass defect detection and classification in weld radiographic images using geometric and texture features. Expert Syst. Appl. 2010. Vol. 37, Iss. 12. pp. 7606–7614. DOI: 10.1016/j.eswa.2010.04.082
9. Kazantsev I., Lemahieu I., Salov G. I., Denys R. Statistical detection of defects in radiographic images in nondestructive testing. Signal Proces. 2002. Vol. 82. pp. 791–801. DOI: 10.1016/S0165-1684(02)00158-5
10. Zou Y., Du D., Chang B., Ji L., Pan J. Automatic weld defect detection method based on Kalman filtering for real-time radiographic inspection of spiral pipe. NDT & E International. 2015. Vol. 72. pp. 1–9. DOI: 10.1016/j.ndteint.2015.01.002
11. Emelyanova M. G., Smailova S. S., Baklanova O. E. Detection of surface defects of welded joints during visual inspection by machine vision methods. Kompyuternaya optika. 2023. No. 1. Available at: https://cyberleninka.ru/article/n/obnaruzhenie-poverhnostnyh-defektov-svarnyh-soedineniy-pri-vizualnom-kontrole-metodami-mashinnogo-zreniya (accessed: 19.09.2024).
12. Severstal: Steel Defect Detection. Available at: https://www.kaggle.com/competitions/severstal-steel-defect-detection/overview (accessed: 19.09.2024).
13. Severstal: Steel Defect Detection. Available at: https://www.kaggle.com/competitions/severstal-steel-defect-detection/discussion/114254 (accessed: 19.09.2024).
14. Defect recognition system. Available at: https://statanly.com/home/case/6 (accessed: 19.09.2024).
15. Control of defects in hot metal cutting using computer vision. Available at: https://atomvestnik.ru/2024/04/01/umnaja-sistema-ml-sense-ot-nord-clan-na-predprijatii-rosatoma/ (accessed: 19.09.2024).
16. Ultralytics YOLOv8. Available at: https://docs.ultralytics.com/ru/models/yolov8/ (accessed: 19.09.2024).
17. Zhou M., Lu W., Xia J., Wang Y. Defect detection in steel using a hybrid attention network. Sensors. 2023. Vol. 23. 6982.
18. CVAT Documentation. Available at: https://docs.cvat.ai/docs (accessed: 19.09.2024).
19. Resnet101. Available at: https://pytorch.org/vision/m