sibsutisВестник СибГУТИThe Herald of the Siberian State University of Telecommunications and Information Science1998-6920СибГУТИ10.55648/1998-6920-2025-19-3-19-29sibsutis-936Research ArticleСтатьиРаспознавание дорожных знаков в сложных метеоусловияхRecognition of road signs in difficult weather conditionshttps://orcid.org/0000-0001-5935-9710ЭдельГерман ЕвгеньевичEdelGerman Evgenievich

Аспирант кафедры телевидения и управления, инженер лаборатории телевидения и автоматики (каф.ТУ). 

german.edel99@gmail.comhttps://orcid.org/0000-0002-2293-0511КапустинВячеслав ВалериевичKapustinVyacheslav Valerievich

Доцент, кандидат технических наук, заведующий лабораторией телевизионной автоматики (каф.ТУ). 

peregnun@mail.ruhttps://orcid.org/0009-0002-7703-3081СукотноваМарина ЕвгеньевнаSukutnovaMarina Evgenievna

Магистр, инженер лаборатории телевидения и автоматики, Томский государственный университет систем управления и радиоэлектроники 

msukotnova@mail.ruТомский государственный университет систем управления и радиоэлектроникиРоссияTomsk State University of Control Systems and RadioelectronicsRussian Federation202530092025193Вестник СибГУТИ1929Copyright © Эдель Г.Е., Капустин В.В., Сукотнова М.Е., 20252025Эдель Г.Е., Капустин В.В., Сукотнова М.Е.Edel G.E., Kapustin V.V., Sukutnova M.E.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://vestnik.sibsutis.ru/jour/article/view/936

Рассмотрена проблема обнаружения и распознавания дорожных знаков в сложных метеоусловиях. Предложен алгоритм, основанный на комбинации моделей нейронных сетей и обладающий высокой точностью и устойчивостью при распознавании изображений, полученных в сложных метеоусловиях. Эффективность алгоритма проверена на наборе изображений, перекрываемых дождем и снегом. Результаты экспериментов показали существенное повышение эффективности предложенного алгоритма по сравнению с алгоритмом, не учитывающим влияние погоды.

The problem of detecting and recognizing road signs in difficult weather conditions is considered. An algorithm based on a combination of neural network models and possessing high accuracy and stability in recognizing images obtained in difficult weather conditions is proposed. The effectiveness of the algorithm was tested on a set of images overlapped by rain and snow. The experimental results showed a significant increase in the efficiency of the proposed algorithm compared to the algorithm that does not take into account the influence of the weather.

автоматическое распознаваниекомпьютерное зрениеповышение точностипредобработка изображенийнейронные сетидинамические метеоусловияобнаружение дорожных знаковautomatic recognitioncomputer visionimproved accuracyimage preprocessingneural networksdynamic weather conditionsroad sign detectionReferencesMathias M., Timofte R., Benenson R., Van Gool L. Traffic sign recognition – How far are we from the solution? // The 2013 international joint conference on Neural networks (IJCNN IEEE), 4-9 August 2013 Fairmont Hotel Dallas, Texas, USA. Р. 1–8.Mathias M., Timofte R., Benenson R., Van Gool L. Traffic sign recognition – How far are we from the solution? The 2013 international joint conference on Neural networks (IJCNN IEEE), 4-9 August 2013 Fairmont Hotel Dallas, Texas, USA. pp. 1-8.Сагдуллаев Ю.С. Отображение и фиксация выделенной телеметрической информации в телевизионных изображениях космических // Вопросы радиоэлектроники, Серия техника телевидения, 2018. Вып. 3, - С 101-107.Sagdullaev Y.S. Display and fixation of selected telemetric information in satellite television im-ages. Voprosy Radioelectronics, TV Technology Series, 2018. Issue 3, pp. 101-107.Фахми Ш. С., Еид М. М., Костикова Е. В. [и др.] Метод и алгоритмы обнаружения и распознавания дорожных знаков // Вопросы радиоэлектроники, Серия техника телевидения, 2018. Вып. 3. С. 95-100.Fahmi Sh. S., Eid M. M., Kostikova E. V. et al. Method and algorithms for detecting and recogniz-ing road signs. Questions of radio electronics, TV Technique Series, 2018. Issue 3. pp. 95-100.Levinson J., Askelad J., Becker J., Dolson J. Towards fully autonomous driving: Systems and algorithms // IEEE intelligent vehicles symposium (IV) IEEE, 5-9 June 2011 Baden-Baden, Germany. Р. 163–168.Levinson J., Askelad J., Becker J., Dolson J. Towards fully autonomous driving: Systems and al-gorithms. IEEE intelligent vehicles symposium (IV) IEEE, 5-9 June 2011 Baden-Baden, Germany. pp. 163–168.Timofte R., Zimmermann K., Van Gool L. Multi-view traffic sign detection, recognition, and 3D localization, Machine vision and applications // IEEE Workshop on Applications of Computer Vision, 7-8 December, 2009, Snowbird, UT, USA. Р. 633–647.Timofte R., Zimmermann K., Van Gool L. Multi-view traffic sign detection, recognition, and 3D localization, Machine vision and applications. IEEE Workshop on Applications of Computer Vi-sion, 7-8 December, 2009, Snowbird, UT, USA. pp. 633-647.Sermanet P., LeCun Y., Traffic sign recognition with multi-scale convolutional networks // The 2011 international joint conference on neural networks IEEE, 31 July – 5 August, 2011 Double-tree Hotel San Jose, California, USA. Р. 2809–2813.Sermanet P., LeCun Y., Traffic sign recognition with multi-scale convolutional networks. The 2011 international joint conference on neural networks IEEE, 31 July – 5 August, 2011 Double-tree Hotel San Jose, California, USA. pp. 2809-2813.Houben S., Stallkamp J., Salmen J., Schlipsing M., Igel C. Detection of traffic signs in real-world images: The German Traffic Sign Detection Benchmark // The 2013 international joint conference on neural networks IJCNN IEEE, 4-9 August 2013 Fairmont Hotel Dallas, Texas, USA. P. 1–8.Houben S., Stallkamp J., Salmen J., Schlipsing M., Igel C. Detection of traffic signs in real-world images: The German Traffic Sign Detection Benchmark. The 2013 international joint conference on neural networks IJCNN IEEE, 4-9 August 2013 Fairmont Hotel Dallas, Texas, USA. pp. 1-8.Q. Guo, J. Sun, F. Juefei-Xu [et al] Efficientderain: Learningpixel-wise dilation filtering for high-efficiency single-imagederaining // Conference: Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI 2021). Р. 1–9.Guo Q., Sun J., Juefei-Xu F. et al. Efficientderain: Learningpixel-wise dilation filtering for high-efficiency single-imagederaining. Conference: Thirty-Fifth AAAI Conference on Artificial Intelli-gence (AAAI 2021). pp. 1-9.Fu H., Huang J., Delu Z. [et al] Removing rain from single images via a deep detail network // In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017. Р. 3855–3863.Fu H., Huang J., Delu Z. et al. Removing rain from single images via a deep detail network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017. pp. 3855-3863.Mu Y., Ruiwen N., Chang Z. [et al] Model of VGG-16 for Remote Sensing Image Classification // IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing PP, June 2021. Р. 6916-6922.Mu Y., Ruiwen N., Chang Z. et al. Model of VGG-16 for Remote Sensing Image Classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing PP, June 2021. pp. 6916-6922.Qi Y., Zhang P., The Tracker with Online Training Based on the TLD Algorithm // Proceedings of the 2nd International Conference on Information Technology and Intelligent Transportation Systems (ITITS 2017), 10 June 2017, Xi’an, China. Р. 76-81.Qi Y., Zhang P., The Tracker with Online Training Based on the TLD Algorithm. Proceedings of the 2nd International Conference on Information Technology and Intelligent Transportation Sys-tems (ITITS 2017), 10 June 2017, Xi'an, China. pp. 76-81.Official website with dataset [Электронный ресурс]. URL: https://paperswithcode.com/dataset/snow100k (дата обращения: 20.05.2024).Official website with dataset, available at: https://paperswithcode.com/dataset/snow100k (accessed: 05.20.2024).Kapustin V.V., Zahlebin A.S., Movchan A.K. [et al] Experimental assessment of the distance measurement accuracy using the active-pulse television measuring system and a digital terrain model // Computer Optics 2022; 46(6). Р. 948-954.Kapustin V.V., Zahlebin A.S., Movchan A.K. et al. Experimental assessment of the distance measurement accuracy using the active-pulse television measuring system and a digital terrain model. Computer Optics 2022; 46(6). pp. 948-954.Movhan A.K., Kapustin V.V., Kuryachiy M.I., Movchan E.S. Multi-Area Method of a Depth Map Building with Gain Modulation in Active-Pulse Television Measuring Systems // In 2022 International Siberian Conference on Control and Communications (SIBCON) Р. 1-6.Movhan A.K., Kapustin V.V., Kuryachiy M.I., Movchan E.S. Multi-Area Method of a Depth Map Building with Gain Modulation in Active-Pulse Television Measuring Systems. In 2022 Interna-tional Siberian Conference on Control and Communications (SIBCON) pp. 1-6.Капустин В.В., Mовчан A.К. Многозонные методы формирования карт глубин с использованием активно-импульсных телевизионных измерительных систем // Вопросы радиоэлектроники, Серия техника телевидения, 2023. Вып. 2. С. 44-54.Kapustin V.V., Movchan A.K. Multi-zone methods of forming depth maps using active pulse tele-vision measuring systems. Issues of Radio Electronics, Television Technology Series, 2023. Issue. 2. pp. 44-54.

The authors declare that there are no conflicts of interest present.