APPLICATION OF COMPUTER VISION TECHNOLOGIES IN VISUAL INFORMATION PROCESSING SYSTEMS
Abstract
This paper considers the application of artificial intelligence technologies, in particular computer vision, in visual information processing systems. A comprehensive analysis of neural network approaches to solving computer vision problems is carried out, including systematization of key types of problems: image classification, object detection and semantic segmentation. The architectural principles of convolutional neural networks are studied in detail with an emphasis on the mechanisms of spatial feature extraction through convolutional layers, optimization of data representation through pooling operations and feature transformation in fully connected layers. Particular attention is paid to the evolution of object detection methods, where the problem of model selection is considered as an extension of classification due to the integration of spatial coordinate regression, and an assessment of the effectiveness of detectors is carried out based on the IoU, Precision, Recall and F1-score metrics, demonstrating a fundamental trade-off between localization accuracy and processing speed. The YOLOv7 algorithm is presented as an optimal solution for real-time systems. Its architecture is based on splitting the input image into a grid of S×S cells with direct prediction of the bounding box parameters (center coordinates, width, height) and class probabilities for each cell, as well as the use of specialized layers (SPP, PANet) for multi-scale feature aggregation. The structure of the neural network confirms the effectiveness of the approach used, which ensures high performance without critically reducing accuracy in strategically important applications of video surveillance, autonomous systems, and augmented reality. A comparative study of one-stage and two-stage detectors was conducted with an assessment of their performance by key metrics. Particular attention is paid to the practical aspects of using computer vision technologies in real visual information processing systems.
References
1. Nikolenko S., Kadurin A., Arkhipel'skaya E. Glubokoe obuchenie. Pogruzhenie v mir neyronnykh setey [Deep learning. Immersion in the world of neural networks]. 2nd ed. Saint Petersburg: Piter, 2023, 576 p.
2. Davies E.R., Turk M.A. Advanced Methods and Deep Learning in Computer Vision. Academic Press, 2022, 690 p.
3. Géron A. Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow. O'Reilly Media, 2022, 870 p.
4. Goryachkin B.S., Kitov M.A. Komp'yuternoe zrenie: sovremennye tendentsii [Computer vision: modern trends], Tsifrovaya obrabotka signalov [Digital signal processing], 2023, No. 1, pp. 45-62.
5. Bovik A.C. Handbook of Image and Video Processing. Academic Press, 2023, 1200 p.
6. Kochanov D.N. Tendentsii razvitiya komp'yuternogo zreniya na osnove glubokogo obucheniya [Trends in the development of computer vision based on deep learning], Iskusstvennyy intellekt v tekhnicheskikh sistemakh: Sb. trudov XII Mezhdunarodnoy nauchno-tekhnicheskoy konferentsii [Artificial intelligence in technical systems: Proceedings of the XII International scientific and technical conference]. Moscow: MGTU im. N.E. Baumana, 2023, pp. 112-119.
7. Zhang J., Li C., Wan X. Real-Time Safety Helmet Detection in Complex Construction Environments, IEEE Transactions on Industrial Informatics, 2023, Vol. 19 (10), pp. 10034-10043.
8. Redmon J., Farhadi A. YOLOv7: An Incremental Improvement, arXiv:1804.02767 [cs.CV], 2023. Available at: https://github.com/ultralytics/ultralytics.
9. Lebedev V.B., Lebedev O.B. Kompozitnye mnogoagentnye sistemy dlya raspoznavaniya izobrazheniy v real'nom vremeni [Composite multi-agent systems for image recognition in real time], Informatika i sis-temy upravleniya [Computer Science and Control Systems], 2022, No. 3 (73), pp. 77-89.
10. Dudarev D.S., Dudarev K.S., Motaylenko L.V. Computer Vision: A Retrospective Analysis of Evolution and Impact, IEEE Access, 2024, Vol. 12, pp. 11245-11260.
11. Padilla R., Passos W.L., Dias T.L. A Comparative Analysis of Object Detection Metrics with a Com-panion Open-Source Toolkit, Electronics, 2021, Vol. 10 (3), pp. 279-284.
12. Dyachenko R.A., Dovgal V.V., Gura D.A. Comparative Analysis of YOLOv7 and U-Net for Remote Sensing Image Segmentation, IEEE Geoscience and Remote Sensing Letters, 2024, Vol. 21, pp. 125-142.
13. Wang Z., Wang P., Li Y. Deep Learning for Face Recognition in Unconstrained Environments: A Sur-vey, ACM Computing Surveys, 2023, Vol. 55 (9), Article 188.
14. Wang C.-Y., Bochkovskiy A., Liao H.-Y. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023, pp. 7464-7475.
15. Vaswani A., Shazeer N., Parmar N. Attention Is All You Need, Advances in Neural Information Pro-cessing Systems 30 (NIPS 2017), 2017, pp. 67-84.
16. Liu Y., Sun P., Wergeles N. A Survey and Performance Evaluation of Deep Learning Methods for Small Object Detection, Expert Systems with Applications, 2021, Vol. 172, pp. 357-369.
17. Kaznacheeva A.A., Vlasenko O.M., Epov A.A. Algoritm upravleniya mekhatronnoy stantsiey sortirovki izdeliy s primeneniem sistemy komp'yuternogo zreniya [Algorithm for controlling a mechatronic station for sorting products using a computer vision system], Elektronnyy nauchnyy zhurnal «Inzhenernyy vest-nik Dona» [Electronic scientific journal «Engineering Bulletin of the Don»], 2025, No. 7 (127), pp. 133-143.
18. Chzhen A., Kazari A. Mashinnoe obuchenie. Konstruirovanie priznakov [Machine learning. Feature engineering]. Moscow: Bombora, 2024, 240 p.
19. Nebaba S.G., Markov N.G. Svertochnye neyronnye seti semeystva YOLO dlya mobil'nykh sistem komp'yuternogo zreniya [Convolutional neural networks of the YOLO family for mobile computer vi-sion systems], Komp'yuternye issledovaniya i modelirovanie [Computer Research and Modeling], 2024, No. 3, pp. 615-631.
20. Trubin A.E. i dr. Metodika predobrabotki dannykh mashinnogo obucheniya dlya resheniya zadach komp'yuternogo zreniya [Methodology for preprocessing machine learning data for solving computer vi-sion problems], Prikladnaya Informatika [Applied Informatics], 2022, No. 4, pp. 36-39.
21. Vasil'ev M.E., Shalimov A.S., Savina O.A. Obzor versiy YOLO: odnoetapnaya model' svertochnoy ney-ronnoy seti [Review of YOLO versions: one-stage model of convolutional neural network], Universum: tekhnicheskie nauki: elektronnyy nauchnyy zhurnal [Universum: technical sciences: electronic scientific journal], 2025, No. 6 (135). Available at: https://7universum.com/ru/tech/archive/item/20293.
22. Krasnoperova A.S., Tverdokhlebov A.S., Kartashov A.A., Veber V.I., Kuprits V.Yu. Issledovanie effek-tivnosti primeneniya modeley neyronnykh setey YOLO dlya raspoznavaniya ob"ektov na radiolo-katsionnykh izobrazheniyakh [Efficiency of YOLO neural network models applied for object recognition in radar images], Russian Technological Journal, 2025, 13 (4), pp. 25-36. Available at: https://doi.org/10.32362/2500-316X-2025-13-4-25-36. EDN: WVWVCJ.
