Journal "Software Engineering"
a journal on theoretical and applied science and technology
ISSN 2220-3397

Issue N5 2026 year

DOI: 10.17587/prin.17.280-292
Architecture and Implementation of a Software System for Wheel Identification by Markings Using Incremental Learning
A. V. Mareev, Postgraduate Student, a.v.mareev@yandex.ru, A. A. Orlov, D. Sc. (Eng.), alexeyalexorlov@gmail.com, I. D. Vereshchagin, Student, ilay.net1@gmail.com, Vladimir State University named after Alexander and Nikolay Stoletovs, Vladimir, 600000, Russian Federation
Corresponding author: Andrey V. Mareev, Postgraduate Student, Vladimir State University named after Alexander and Nikolay Stoletovs, Vladimir, 600000, Russian Federation, E-mail: a.v.mareev@yandex.ru
Received on September 02, 2025
Accepted on November 18, 2025

This article addresses the task of ensuring end-to-end traceability in railway wheel production lines using modern computer vision methods. The aim of the work is to design and develop a software system for the automatic identification of wheels based on their markings, which is a critical element for the digitalization of manufacturing processes. The algorithm proposed by the authors is a multi-stage video processing pipeline based on the YOLOv11 family of neural network architecture and the ByteTrack object tracker. The pipeline involves the sequential execution of the following operations: detection and tracking of wheels in frames, precise localization of the marking area, its geometric alignment and cropping, segmentation and classification of individual characters, and the formation of a final marking based on a consensus approach that aggregates results across a series of frames. A key feature of the system, distinguishing it from traditional solutions, is the implementation of an incremental fine-tuning module. This module allows for continuous improvement of recognition accuracy during operation by adapting the model to new marking types without the need for complete and resource-intensive retraining on the entire dataset. The article presents the results of a comprehensive experimental study confirming the system's high efficiency. A detection performance of mAP@0.5 = 0.995 was achieved on test data, with the system demonstrating robustness to typical industrial noise (contamination, glare, lighting variations). The system's performance reaches up to 35 frames per second on high-performance hardware. The modular architecture of the developed standalone application and its user interface are described in detail. The interface ensures convenient operator interaction with the system, including capabilities for manual correction and initiating the incremental fine-tuning process. The practical significance of the work lies in creating a universally applicable tool, robust to industrial noise, applicable not only in the railway industry but also in other fields of mechanical engineering for solving automated identification tasks for large-scale products.

Keywords: computer vision system, neural networks, feature extraction, recognition, symbols, marking, incremental learning, accuracy, railway wheels, YOLOv11
pp. 280—292
For citation:
Mareev A. V., Orlov A. A., Vereshchagin I. D. Architecture and Implementation of a Software System for Wheel Identification by Markings Using Incremental Learning, Programmnaya Ingeneria, 2026, vol. 17, no. 5, pp. 280—292. DOI: 10.17587/prin.17.280-292 (in Russian).
This work was conducted with the support of the grant from the Russian Science Foundation, grant number 24-21-00386. You can find further details about the grant at the following link: https://rscf.ru/en/project/24-21-00386/.
References:
  1. Kralicek J., Matas J., Busta M. Care Label Recognition, 2019 International Conference on Document Analysis and Recognition (ICDAR), Sydney, NSW, Australia, 2019, pp. 959—966. DOI: 10.1109/ICDAR.2019.00158.
  2. Shekar Goud D., Vigneshwari M. et al. Text Localization and Recognition from Natural Scene Images using AI, 2022 International Conference on Automation, Computing and Renewable Systems (ICACRS), Pudukkottai, India, 2022, pp. 1153—1158. DOI: 10.1109/ICACRS55517.2022.10029220.
  3. Tourani A., Shahbahrami A., Soroori S. et al. A Robust Deep Learning Approach for Automatic Iranian Vehicle License Plate De­tection and Recognition for Surveillance Systems, IEEE Access, 2020, vol. 8, pp. 201317—201330. DOI: 10.1109/ACCESS.2020.3035992.
  4. Xiang Z., Wu H., Zhou D. Recognition of characters on curved metal workpiece surfaces based on multi-exposure image fusion and deep neural networks, Neurocomputing, 2022, vol. 512, pp. 117—129. DOI: 10.1016/j.neucom.2022.09.022.
  5. Xiang Z., Huang Y. A Metal Character Enhancement Method based on Conditional Generative Adversarial Networks, 2022 International Conference on Machine Vision, Automatic Identification and Detection (MVAID 2022), 2022, vol. 2284, article 012003. DOI: 10.1088/1742-6596/2284/1/012003.
  6. Liu S., Li Y., Zhao A., Liu K., Wang J. A brake disc character recognition method based on photometric stereo and surface curvature extraction, 2019 Chinese Automation Congress (CAC), Hangzhou, China, 2019, pp. 3324—3329. DOI: 10.1109/CAC48633.2019.8996247.
  7. Srivastava S. Comparative analysis of deep learning image detection algorithms, Journal of Big Data, 2021, vol. 8, article 66. DOI: 10.1186/s40537-021-00434-w.
  8. Erokhin D. Y., Ershov M. D. Modern convolutional neural networks for object detection and recognition, Digital Signal Processing, 2018, no. 3, pp. 64—69 (in Russian).
  9. Kotov N. V., Khryashchev V. V., Tikhomirov A. A. Investigation of YOLO neural network architecture for cecum detection in colonoscopy, Digital Signal Processing and Its Applications — DSPA— 2024, Moscow, Russian scientific and technical society of radio engineering, electronics and communication of name A. S. Popov, 2024, pp. 151—155 (in Russian).
  10. Zhgutov P. E., Muraviev V. S. Comparative study of neural network algorithms for traffic density estimation, Digital Signal Processing and Its Applications DSPA—2023, Moscow, Russian scientific and technical society of radio engineering, electronics and com­munication of name A. S. Popov, 2023, pp. 205—209 (in Russian).
  11. Kirnos V. P., Kokovkina V. A. Development of a simultaneous localization and mapping system based on neural networks, Digital Signal Processing and Its Applications DSPA—2023, Moscow, Russian scientific and technical society of radio engineering, elec­tronics and communication of name A. S. Popov, 2023, pp. 363—366 (in Russian).
  12. Potapov E., Artamonov K., Sidorova E. YOLOv7 convolutional neural network for object detection in images for a spherical robot, Digital Signal Processing and Its Applications DSPA—2023, Moscow, Russian scientific and technical society of radio engineering, electronics and communication of name A. S. Popov, 2023, pp. 367—371 (in Russian).
  13. Vasilev D. A., Levkin T. V., Skonnikov P. N., Trofimov D. V. A modification of YOLO object detector for real-time implementation on FPGA, Digital Signal Processing, 2022, no. 3, pp. 77—79 (in Russian).
  14. Khryashchev V. V., Priorov A. L., Lebedev A. A. et al. Detection of polyps on colonoscopic images using algorithms based on the YOLO neural network architecture, Digital Signal Processing, 2023, no. 3, pp. 27—31 (in Russian).
  15. Jegham N., Koh C. Y., Abdelatti M., Hendawi A. Evaluating the Evolution of YOLO (You Only Look Once) Models: A Comprehensive Benchmark Study of YOLO11 and Its Predecessors, arXiv, 2024, DOI: 10.48550/arXiv.2411.00201.
  16. Sapkota R., Karkee M. Comparing YOLO11 and YOLOv8 for instance segmentation of occluded and non-occluded immature green fruits in complex orchard environment, arXiv, 2024, DOI: 10.48550/arXiv.2410.19869.
  17. Mareev A. V., Orlov A. A., Vereshchagin I. D. Adaptive Recognition of Markings on Railway Wheels using Incremental Learning With YOLOv11, 2025 International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM), Sochi, Russian Federation, 2025, IEEE, 2025, pp. 789—793. DOI: 10.1109/ICIEAM65163.2025.11028540.
  18. Mareev A. V., Orlov A. A., Vereshchagin I. D. Investigation of yolo neural network architecture for identification of railway wheels in video stream with incremental learning. Digital Signal Processing and Its Applications — DSPA — 2025, Moscow, Russian scientific and technical society of radio engineering, electronics and communication of name A. S. Popov, 2025, pp. 161—166 (in Russian).
  19. Cubuk E. D., Zoph B., Shlens J., Le Q. V. RandAugment: Practical automated data augmentation with a reduced search space, arXiv, 2019. DOI: 10.48550/arXiv.1909.13719.
  20. Bewley A., Ge Z., Ott L. et al. Simple Online and Realtime Tracking, 2016, IEEE International Conference on Image Processing (ICIP), 2016, pp. 3464—3468. DOI: 10.1109/ICIP.2016.7533003.