Detection and classification of dangerous maneuvers based on traffic camera recordings
https://doi.org/10.26518/2071-7296-2025-22-6-916-927
EDN: RGQURR
Abstract
Introduction. Traffic accidents are a leading cause of death. In 2024, 19,940 people were killed in accidents in the EU, while in Russia this number amounts to 14,400 people. A significant proportion of accidents are associated with dangerous maneuvers: abrupt lane changes, overtaking, emergency braking, and running red lights. Traditional monitoring methods are limited by cost and scalability. The goal of this research is to develop a system for automatic detection and classification of dangerous maneuvers based on video data obtained with YOLOv8 and Deep SORT.
Materials and Methods. A system consisting of four modules is proposed: a modified YOLOv8n (with a P2 layer, LW_C2f, Wise-IoU) for vehicle detection; an optimized Deep SORT for tracking; trajectory analysis with camera calibration; maneuver classification based on acceleration thresholds (0.35g for lane change, 0.30g for braking), lane marking crossing, and traffic light status (YOLOv8). The system was trained on 45,000 vehicle images and on 20,000 images for re-ID.
Results. Testing of 150 hours of video (various conditions) has shown the following results: mAP vehicle detection – 92.7%, MOTA tracking – 86.3%, maneuver classification accuracy – 89.3% (F1: lane change – 89.4%, braking – 89.7%, red light – 85.2%) at 28 FPS on RTX 3070. Latency – 0.12 s.
Discussion and conclusions. The system under investigation is superior to similar systems in speed and maneuver coverage, and is suitable for ITS applications. Limitations refer to decreased accuracy in foggy or rainy weather. Further research can lead to expanding the class range, edge computing, and risk prediction. The system implementation will reduce accidents and results in automated traffic enforcement.
Keywords
computer vision,
detection of dangerous maneuvers,
YOLOv8,
Deep SORT,
road safety,
video analytics,
neural networks,
classification of traffic violations
Supporting agencies: This research was completed under the Priority 2030 Program at the Belgorod State Technological University named after V.G. Shukhov. The equipment of High Technology Center at BSTU has been used.
About the Authors
A. N. Novikov
Orel State University named after I.S. Turgenev
Russian Federation
Russian Federation
Novikov Alexander N. – Doctor of Technical Sciences, Professor, Head of the Department of Machine Service and Repair
Scopus ID: 57225227480, Author ID: 143921, Researcher ID: M-4302-2017
77, Moskovskaya str., Orel, 302002
L. E. Kushchenko
Belgorod State Technological University named after V.G. Shukhov
Russian Federation
Russian Federation
Kushchenko Liliya E. – Doctor of Technical Sciences, Associate Professor, Professor of the Department of Vehicle Operation and Traffic Management
Scopus ID: 57193997889, Author ID: 746460
46 Kostyukova str., Belgorod, 308012
S. V. Kushchenko
Belgorod State Technological University named after V.G. Shukhov
Russian Federation
Russian Federation
Kushchenko Sergey V. – Candidate of Technical Sciences, Associate Professor, Associate Professor of the Department of Vehicle Operation and Traffic Management
Scopus ID: 57193997343, Author ID: 970275
46 Kostyukova str., Belgorod, 308012
I. A. Ulinets
Belgorod State Technological University named after V.G. Shukhov
Russian Federation
Russian Federation
Ulinets Iosif A. – Postgraduate Student
Author ID: 1272489
6 Kostyukova str., Belgorod, 308012
References
1. Eremin S.V., Kushchenko L.E., Kushchenko S.V., Novikov A.N. Improving Road Traffic Safety in Urban Agglomerations: Monograph. Belgorod: Belgorod State Technological University named after V. G. Shukhov, 2024. 158 pp. (In Russ.)
2. Road Safety Statistics for 2024: Progress continues amid persistent challenges. European Commission, 2025. URL: https://transport.ec.europa.eu/news-events/news/road-safety-statistics-2024-progress-continues-amid-persistent-challeng-es-2025-10-17_en (accessed: October 22, 2025).
3. Novikov A.N., Kushchenko L.E., Kushchenko S.V., Novikov I.A. Analysis of existing methods for assessing the probability of road traffic accidents on sections of the urban street-road network. Bulletin of Civil Engineers. 2021; no. 2 (85): 222–231. (In Russ.)
4. Novikov I.A., Kravchenko A.A., Shevtsova A.G., Vasilieva V.V. Scientific and methodological approach to reducing road accidents in the Russian Federation. World of Transport and Technological Machines. 2019; no. 3 (66): 58–64. (In Russ.) Https://doi.org/10.33979/2073-7432-2019-66-3-58-64
5. Kushchenko L.E., Kushchenko S.V., Kravchenko A.A., Shatova J.S. Improving traffic safety and accident prediction at pedestrian crossings. IOP Conference Series: Materials Science and Engineering, 2020. vol. 913:042060. Https://doi.org/10.1088/1757-899X/913/4/042060.
6. Kushchenko L., Kushchenko S., Novikov A., Koroleva L. The statistical assessment of the traffic situation based on sample data of traffic accidents in the urban agglomeration. Journal of Applied Engineering Science. 2023; vol. 21, no. 4. Https://doi.org/10.5937/jaes0-42852
7. Zyryanov V.V., Kocherga V.G., Pozdnyakov M.N. Modern approaches to developing comprehensive traffic management schemes. Transport of the Russian Federation. 2011; no. 1: 28–33. (In Russ.)
8. Petrov A.I., Evtyukov S.A., Kolesov V.I. New approaches to road traffic safety management: the paradigm of organized safety assurance processes. World of Transport and Technological Machines. 2019; no. 3 (66): 65–74. (In Russ.) Https://doi.org/10.33979/2073-7432-2019-66-3-58-64
9. Debbarma T., et al. Prediction of Dangerous Driving Behaviour Based on Computer Vision and Machine Learning. Procedia Computer Science. 2024; vol. 235: 783–792. Https://doi.org/10.1016/j.procs.2024.04.178
10. Zhou T., Zhang X., Chen H. A. Dangerous Driving Behavior Detection Method Based on Improved YOLOv8s. Engineering Letters. 2025; vol. 33(3): 721– 731.
11. Luo Z., Bi Y., Yang, X., Li Y., Yu S., Wu M., Ye Q. Enhanced YOLOv5s + DeepSORT method for highway vehicle speed detection and multi-sensor verification. Frontiers in Physics. 2024; vol. 12, Article 1371320. Https://doi.org/10.3389/fphy.2024.1371320
12. Ong C.S. et al. Traffic Violation Detection Using Computer Vision Techniques. Journal of Imaging. 2024; vol. 8(3): 1264–1278. Https://doi.org/10.62527/joiv.8.3-2.2941
13. Pramanik A. et al. A real-time video surveillance system for traffic pre-events detection. Accident Analysis & Prevention. 2021; vol. 154, Article 106068. Https://doi.org/10.1016/j.aap.2021.106068
14. Han L., et al. Transformer-based modeling of abnormal driving events for freeway crash risk evaluation. Transportation Research Part C: Emerging Technologies. 2024; vol. 156, Article 104481. Https://doi.org/10.1016/j.trc.2024.104481
15. Wei Z., et al. Vision-Based Lane-Changing Behavior Detection Using Deep Learning. IEEE Transactions on Intelligent Vehicles. 2019; vol. 4(3): 457– 467. Https://doi.org/10.1109/TIV.2019.2934567
16. Li R. et al. YOLO-SGC: A Dangerous Driving Behavior Detection Method Based on Enhanced YOLOv8. IEEE Transactions on Intelligent Transportation Systems. 2024; Https://doi.org/10.1109/TITS.2024.3456789
17. Gay L.E., Shutov A.I., Volya P.A., Kushchenko S.V. Congestion phenomena: opportunities for prevention. Bulletin of BSTU named after V. G. Shukhov. 2013; no. 3: 166–169. (In Russ.)
18. Wang G. et al. Traffic sign detection method based on improved YOLOv8. Scientific Reports. 2025; vol. 15, Article 3792. Https://doi.org/10.1038/s41598-025-03792-0
19. Dou H., Chen S., Xu F., Liu Y., Zhao H. Analysis of vehicle and pedestrian detection effects of improved YOLOv8 model in drone-assisted urban traffic monitoring system. PLOS ONE. 2025; vol. 20(3). Article e0314817. Https://doi.org/10.1371/journal.pone.0314817
20. Hassan O.F. et al. Real-time driver drowsiness detection using transformer architecture and transfer learning. Scientific Reports. 2025; vol. 15, Article 2111. Https://doi.org/10.1038/s41598-025-02111-x
21. Zyryanov V.V., Kocherga V.G., Pozdnyakov M.N. Modern approaches to developing comprehensive traffic management schemes. Transport of the Russian Federation. 2011; no. 1: 28–33. (In Russ.)
22. Wojke N., Bewley A., Paulus D. Simple online and realtime tracking with a deep association metric. In: 2017 IEEE International Conference on Image Processing (ICIP). pp. 3645–3649. Https://doi.org/10.1109/ICIP.2017.8296962
23. Pechatnova E.V., Kiryushin I.I., Nagorny N.N. Factors of road traffic accident rate in the suburban area; Bulletin of the Siberian State Automobile and Highway University (SibADI). 2024; vol. 21, no. 5 (99). (In Russ.)
Review
For citations:
Novikov A.N., Kushchenko L.E., Kushchenko S.V., Ulinets I.A. Detection and classification of dangerous maneuvers based on traffic camera recordings. The Russian Automobile and Highway Industry Journal. 2025;22(6):916-927. (In Russ.) https://doi.org/10.26518/2071-7296-2025-22-6-916-927. EDN: RGQURR
Views:
10
JATS XML
Email this article 