DOI: 10.17587/prin.14.513-519

Neural Network Technology for Traffic Flow Analysis in Automated Traffic Control Systems

R. M. Khusainov, Postgraduate Student, rumil_husainov98@mail.ru, N. G. Talipov, PhD (technical science), Associate Professor, nafis.talipov@mail.ru, A. S. Katasev, Dr. Sci. (technical science), Professor, ASKatasev@kai.ru, D. V. Shalaeva, Master Student, dvshalaeva@bk.ru, Kazan National Research Technical University named after A. N. Tupolev, Kazan, 420111, Russian Federation

Corresponding author: Nafis G. Talipov, PhD (technical science), Associate Professor, Kazan National Research Technical University named after A. N. Tupolev, Kazan, 420111, Russian Federation, E-mail: nafis.talipov@mail.ru

The article presents the results of solving the problem of developing a neural network technology for analyzing traffic flows in automated traffic control systems. Methods for analyzing traffic flows using various technologies, existing systems for analyzing traffic flows based on neural network technologies and technologies for detecting and tracking objects on a video stream in order to ensure traffic safety are described. The algorithms used in the operation of neural network technology are described, including the stages of object detection; tracking of detected objects; identifying incidents; automatic collection of information from video streams; collecting statistics. According to the results of approbation (testing) of the implemented neural network technology on own (prepared) and used video data downloaded from the Internet, the reliability of the results of the study of video frames (correct recognition of traffic flow objects) in the neural network traffic control system was 85—90 %. Errors occurred in the presence of a large number of objects on a video frames and poor quality of the video stream. When monitoring the parameters of the traffic flow, the collection, analysis and provision of aggregated data on the traffic situation on the road section were ensured. Automatic collection of information from video files is implemented up to the file time without refer­ence to real time. The statistical data obtained during the study can be used in automated traffic control systems to analyse traffic flows and in other control and decision-making systems to increase the responsiveness to incidents. The results obtained can be applied in traffic flow studies to obtain accurate statistics and forecast anomalies and incidents (traffic accidents, traffic jams) in order to ensure traffic safety.

Keywords: traffic flows, object recognition, object detection, incident, neural network, traffic accidents, vehicle, traffic, video stream, neural network technology

pp. 513–519

For citation:
Khusainov R. M., Talipov N. G., Katasev A. S., Shalaeva D. V. Neural Network Technology for Traffic Flow Analysis in Automated Traffic Control Systems, Programmnaya Ingeneria, 2023, vol. 14, no. 10, pp. 513—519. DOI: 10.17587/prin.14.513-519 (in Russian).

References:

• Indicators of the state of road safety, available at: http://stat. gibdd.ru/ (date of access 04.07.2023).
• Yermishina E. V., Sharypova T. N. Critical information infrastructure in the intelligent transport system, Colloquim-Journal, 2022, no. 2-1 (125), pp. 20—21 (in Russian).
• Sokolyansky V. V, Bulatov A. B., Bulatov R. B., Dudayev B. V. Intelligent transport systems in urban, Voprosy ekonomicheskikh nauk, 2015, no. 4 (74), pp. 75—81 (in Russian).
• Nizyaeva Yu. D., Slobodchikov N. A. Experience in implementing intelligent transport systems in urban infrastructure, Sistemnyy analiz i logistika. 2022, no. 2 (32), pp. 50—55. DOI: 10.31799/2077-5687-2022-2-50-55 (in Russian).
• Intelligent accident prevention system, available at: https:// iss.ru/products/securos-soffit (date of access 26.07.2023).
• The Russian AI system Crash AI will warn of the risk of an accident, available at: https://3dnews.ru/981314 (date of access 26.07.2023).
• ASUDD "Megapolis", available at: https://asudd.pro/prod-ucts/asudd/ (date of access 26.07.2023).
• Open source software: why is it especially dangerous today and how to protect yourself, available at: https://www.tadviser.ru/ index.php (date of access 26.07.2023).
• The weak link of Open Source, available at: https://in-nostage-group.ru/press/blog/review/slaboe-zveno-open-source (date of access 26.07.2023).
• Lecun Y., Boser B., Denker J. S., Henderson D., Howard R. E., Hubbard W., Jackel L. D. Backpropagation applied to handwritten zip code recognition, Neural Computation, 1989. pp. 541—551.
• Shleymovich M. P., Dagaeva M. V., Katasev A. S. et al. The analysis of images in control systems of unmanned automobiles on the base of energy features model, Computer Research and Modeling, 2018, no. 10 (3), pp. 369—376. DOI: 10.20537/2076-7633-2018-10-3-369-376 (in Russian).
• Khusainov R. M., Talipov N. G. Analysis of algorithms and systems for recognizing traffic signs, Vestnik tekhnologicheskogo universiteta, 2022, vol. 25, no. 3, pp. 72—77 (in Russian).
• Druki A. A. The use of convolutional neural networks for the extraction and recognition of license plates in images with a complex background, Izvestiya Tomskogo politekhnicheskogo universiteta, 2014, vol. 324, no. 5, pp. 85—92 (in Russian).
• Homonenko A. D., Yakovlev E. L. Substantiation of the architecture of a convolutional neural network for autonomous recognition of objects in images by an onboard computer system, Naukoyemkiye tekhnologii v kosmicheskikh issledovaniyakh Zemli, 2018, vol. 10, no. 6, pp. 86—93 (in Russian).
• Bokov P. A., Kravcheniya P. D. Experimental analysis of the accuracy and performance of varieties of Yolo architectures for computer vision tasks, Programmnyye produktyi sistemy, 2020, no. 4, pp. 635—640. DOI: 10.15827/0236-235X.132.635-640 (in Russian).
• Rudenkyy A. O. Application of architecture to detect viola­tions of rules, stopping, parking vehicles in the wrong place and driving on sidewalks, Mezhdunarodnaya nauchno-tekhnicheskaya konferentsiya molodykh uchenykh BGTU named V. G. Shukhova, Materialy konferentsii, 2021, pp. 3779—3788 (in Russian).
• Schneider A. S. Detection of objects in images on the example of the Yolo model, Trudy molodykh uchenykh Altay-skogo gosudarstvennogo universiteta, 2018, no. 15, pp. 358—361 (in Russian).
• Jung H. Choi M.-K., Jung J. et al. ResNet-Based Vehicle Classificaton an Localization in Traffic Surveillance Systems, IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2017, pp. 934—940. DOI: 10.1109/CVPRW.2017.129.
• Bagirov M. B. Development and research of algorithms for tracking objects on a video stream, Informatsionnyye sistemy i tekhnologii. Sbornik materialov XXVIII Mezhdunarodnoy nauchno-tekhnicheskoy konferentsii, 2022, pp. 350—360 (in Russian).
• Wojke N., Bewley A., Paulus D. Simple Online and Realtime Tracking with a Deep Association Metric, IEEE International Conference on Image Processing, 2017, pp. 1—5. DOI: 10.1109/ ICIP.2017.8296962.
• Anaconda official site, available at: https://www.anaconda. com/products/individual (date of access 04.07.2023).
• Official site of the developer Joseph Chet Redmon, available at: https://pjreddie.com/media/files/yolov3.weights (date of access 04.07.2023).
• Networks. Google-Disk, available at: https://drive.google.com/drive/folders/1m2ebLHB2JThZC8vWGDYEKGsevLssSkjo (date of access 04.07.2023).
• Developer's official github repository, available at: https:// github.com/nwojke/deep_sort (date of access 04.07.2023).
• Developer's official github repository, available at: https:// github.com/qqwweee/keras-yolo3 (date of access 04.07.2023).