Araştırma Makalesi
BibTex RIS Kaynak Göster

Kentsel Trafikte Araç Kuyruğu Durumlarının Derin Öğrenme ile Gerçek Zamanlı Tespiti

Yıl 2025, Cilt: 12 Sayı: 3, 356 - 364, 30.09.2025
https://doi.org/10.31202/ecjse.1755333

Öz

Sinyalize kavşaklarda trafik sıkışıklığı ve araç kuyruğu oluşumu, modern kentsel ulaşım sistemlerinde önemli zorluklar arasında yer almakta olup, etkili trafik yönetimi için doğru ve gerçek zamanlı tespit yöntemleri gerektirmektedir. Bu çalışma, standart trafik kamerası görüntülerini kullanarak You Only Look Once (YOLO) nesne tespiti ile Simple Online Real-time Tracking (SORT) algoritmalarını entegre eden, gerçek zamanlı araç kuyruk durumu sınıflandırmasına yönelik derin öğrenme tabanlı bir yaklaşım sunmaktadır. Önerilen sistem; çok sınıflı araç sınıflandırması, benzersiz kimlik ataması ile gerçek zamanlı araç takibi ve kamera kalibrasyon teknikleri aracılığıyla hız tahmini gibi işlevleri gerçekleştirmektedir. Sistem, farklı video senaryoları arasında ortalama 16.42 FPS işlem hızı elde etmektedir. Geliştirilen kapsamlı kuyruk durumu tespit yöntemi, tespit edilen araçların ortalama hızlarına dayalı olarak trafik durumunu üç kategoriye ayırmaktadır: Yoğun trafik, istikrarlı akış ve serbest akış. Motosikletler ve bisikletler, farklı trafik davranış özellikleri nedeniyle analiz dışında bırakılmıştır. Hem yüksek hem de düşük çözünürlüklü çeşitli test veri kümelerinde yapılan deneysel doğrulama, tüm araç sınıflarında sağlam bir tespit performansı göstermektedir. Hız tahmin doğruluğu, belirli trafik şeritlerinde yapılan araç sayımı ve takibiyle doğrulanmış olup, %89 ile %99 arasında değişmektedir. Araç sayım doğruluğu ise farklı senaryolarda %78.57 ile %100 arasında değişmektedir. Bu sistem, mevcut trafik gözetim altyapısını kullanarak geleneksel sensör tabanlı yöntemlere kıyasla maliyet etkin bir alternatif sunmakta ve akıllı ulaşım sistemlerinde yaygın uygulama için uygun bir çözüm oluşturmaktadır. Elde edilen sonuçlar, önerilen yaklaşımın, yoğun trafik sıkışıklığından serbest akış koşullarına kadar çeşitli trafik senaryolarında kuyruk durumlarını gerçek zamanlı olarak başarıyla tespit ettiğini göstermektedir. Bu araştırma, entegre nesne tespiti ve izleme algoritmalarının pratik etkinliğini ortaya koyarak, bilgisayarla görme tabanlı trafik izleme alanına katkı sağlamakta ve trafik akışının iyileştirilmesi ile sıkışıklık yönetimine yönelik önemli bir adım sunmaktadır.

Proje Numarası

2023/AP/0002

Kaynakça

  • [1] S. Lee, K. Xie, D. Ngoduy, and M. Keyvan-Ekbatani, ‘‘An advanced deep learning approach to real-time estimation of lane-based queue lengths at a signalized junction,’’ Transportation research part C: emerging technologies, vol. 109, pp. 117–136, 2019.
  • [2] R. Rahman and S. Hasan, ‘‘Real-time signal queue length prediction using long short-term memory neural network,’’ Neural Computing and Applications, vol. 33, pp. 3311–3324, 2021.
  • [3] M. Umair, M. U. Farooq, R. H. Raza, Q. Chen, and B. Abdulhai, ‘‘Efficient video-based vehicle queue length estimation using computer vision and deep learning for an urban traffic scenario,’’ Processes, vol. 9, no. 10, p. 1786, 2021.
  • [4] J. Wu, H. Xu, Y. Zhang, Y. Tian, and X. Song, ‘‘Real-time queue length detection with roadside lidar data,’’ Sensors, vol. 20, no. 8, p. 2342, 2020.
  • [5] Y. Zhao, J. Zheng,W.Wong, X.Wang, Y. Meng, and H. X. Liu, ‘‘Various methods for queue length and traffic volume estimation using probe vehicle trajectories,’’ Transportation Research Part C: Emerging Technologies, vol. 107, pp. 70–91, 2019.
  • [6] G. Comert, T. Amdeberhan, N. Begashaw, N. G. Medhin, and M. Chowdhury, ‘‘Simple analytical models for estimating the queue lengths from probe vehicles at traffic signals: A combinatorial approach for nonparametric models,’’ Expert Systems with Applications, vol. 252, p. 124076, 2024.
  • [7] Q. Zhou, R. Mohammadi, W. Zhao, K. Zhang, L. Zhang, Y. Wang, C. Roncoli, and S. Hu, ‘‘Queue profile identification at signalized intersections with highresolution data from drones,’’ in 2021 7th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS), pp. 1–6, IEEE, 2021.
  • [8] S. Jayatilleke, V. Wickramasinghe, and N. Amarasingha, ‘‘Introduction of a simple estimation method for lane-based queue lengths with lane-changing movements,’’ Journal of The Institution of Engineers (India): Series A, vol. 104, no. 1, pp. 143–153, 2023.
  • [9] M. Y. Arafat, A. S. M. Khairuddin, U. Khairuddin, and R. Paramesran, ‘‘Systematic review on vehicular licence plate recognition framework in intelligent transport systems,’’ IET intelligent transport systems, vol. 13, no. 5, pp. 745–755, 2019.
  • [10] D. Liu, C. An, M. Yasir, J. Lu, and J. Xia, ‘‘A machine learning based method for real-time queue length estimation using license plate recognition and gps trajectory data,’’ KSCE Journal of Civil Engineering, vol. 26, no. 5, pp. 2408–2419, 2022.
  • [11] X. Zhan, R. Li, and S. V. Ukkusuri, ‘‘Lane-based real-time queue length estimation using license plate recognition data,’’ Transportation Research Part C: Emerging Technologies, vol. 57, pp. 85–102, 2015.
  • [12] P. Pudasaini, A. Karimpour, and Y.-J. Wu, ‘‘Real-time queue length estimation for signalized intersections using single-channel advance detector data,’’ Transportation research record, vol. 2677, no. 7, pp. 144–156, 2023.
  • [13] W. Al Okaishi, A. Zaarane, I. Slimani, I. Atouf, and M. Benrabh, ‘‘A vehicular queue length measurement system in real-time based on ssd network,’’ Transport and Telecommunication, vol. 22, no. 1, pp. 29–38, 2021.
  • [14] Ultralytics, ‘‘Yolov5.’’ https://github.com/ultralytics/yolov5?tab=readme-ov-file. (Accessed: 2023-02-10).
  • [15] A. Battal, Y. Avci, and A. Tuncer, ‘‘Vehicle detection and counting in traffic videos using deep learning,’’ ICENTE’23, p. 272, 2023.
  • [16] M. Boneh, ‘‘Vehicle detection.’’ https://github.com/MaryamBoneh/VehicleDetection/tree/main/Dataset. (Accessed: 2023-02-12).
  • [17] L. Soetanto, ‘‘Vehicle detection.’’ https://www.kaggle.com/datasets/lyensoetanto/vehicle-images-dataset. (Accessed: 2023-02-21).
  • [18] M. TÁRNOK, ‘‘5 vehichles for classification.’’ https://www.kaggle.com/datasets/mrtontrnok/5-vehichles-for-multicategory-classification. (Accessed: 2024-06- 06).
  • [19] A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, ‘‘Simple online and realtime tracking,’’ in 2016 IEEE international conference on image processing (ICIP), pp. 3464–3468, Ieee, 2016.
  • [20] P. K. Thadagoppula and V. Upadhyaya, ‘‘Speed detection using image processing,’’ in 2016 International Conference on Computer, Control, Informatics and its Applications (IC3INA), pp. 11–16, IEEE, 2016.
  • [21] C.-J. Lin, S.-Y. Jeng, and H.-W. Lioa, ‘‘A real-time vehicle counting, speed estimation, and classification system based on virtual detection zone and yolo,’’ Mathematical Problems in Engineering, vol. 2021, no. 1, p. 1577614, 2021.
  • [22] A. M. Santos, C. J. Bastos-Filho, A. M. Maciel, and E. Lima, ‘‘Counting vehicle with high-precision in brazilian roads using yolov3 and deep sort,’’ in 2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), pp. 69–76, IEEE, 2020.
  • [23] Y. Zhao, X. Zhou, X. Xu, Z. Jiang, F. Cheng, J. Tang, and Y. Shen, ‘‘A novel vehicle tracking id switches algorithm for driving recording sensors,’’ Sensors, vol. 20, no. 13, p. 3638, 2020.

Real-Time Detection of Vehicle Queue States in Urban Traffic Using Deep Learning

Yıl 2025, Cilt: 12 Sayı: 3, 356 - 364, 30.09.2025
https://doi.org/10.31202/ecjse.1755333

Öz

Traffic congestion and vehicle queue formation at signalized intersections represent critical challenges in modern urban transportation systems, requiring accurate real-time detection methods for effective traffic management. This study presents a deep learning-based approach for real-time vehicle queue state classification that integrates You Only Look Once (YOLO) object detection with Simple Online Real-time Tracking (SORT) algorithms using standard traffic camera footage. The proposed system performs multi-class vehicle classification, real-time vehicle tracking with unique ID assignment, and speed estimation through camera calibration techniques, achieving 16.42 FPS average processing speed across diverse video scenarios. A comprehensive queue state detection methodology is developed that categorizes traffic conditions into three categories: Heavy traffic, stable flow, and free flow based on the analysis of average speeds of the detected vehicles, excluding motorcycles and bicycles due to their distinct traffic behavior patterns. Experimental validation across several test datasets encompassing both high and low resolutions demonstrates robust vehicle detection performance across all vehicle classes. Speed estimation accuracy ranges from 89% to 99%, validated against vehicle counting and tracking in designated traffic lanes, providing essential data for queue analysis. The system achieves vehicle counting accuracy ranging from 78.57% to 100% across different scenarios. The system offers a cost-effective alternative to traditional sensor-based methods by utilizing existing traffic-surveillance infrastructure, making it suitable for widespread deployment in intelligent transportation systems. Results indicate the proposed approach successfully detects queue states in real-time conditions across diverse traffic scenarios, from heavy congestion to free flow conditions. This research advances computer vision-based traffic monitoring by demonstrating the practical effectiveness of integrated object detection and tracking algorithms, contributing to improved traffic flow optimization and congestion management.

Destekleyen Kurum

Yalova University

Proje Numarası

2023/AP/0002

Teşekkür

This study was supported by the Research Fund of Yalova University.

Kaynakça

  • [1] S. Lee, K. Xie, D. Ngoduy, and M. Keyvan-Ekbatani, ‘‘An advanced deep learning approach to real-time estimation of lane-based queue lengths at a signalized junction,’’ Transportation research part C: emerging technologies, vol. 109, pp. 117–136, 2019.
  • [2] R. Rahman and S. Hasan, ‘‘Real-time signal queue length prediction using long short-term memory neural network,’’ Neural Computing and Applications, vol. 33, pp. 3311–3324, 2021.
  • [3] M. Umair, M. U. Farooq, R. H. Raza, Q. Chen, and B. Abdulhai, ‘‘Efficient video-based vehicle queue length estimation using computer vision and deep learning for an urban traffic scenario,’’ Processes, vol. 9, no. 10, p. 1786, 2021.
  • [4] J. Wu, H. Xu, Y. Zhang, Y. Tian, and X. Song, ‘‘Real-time queue length detection with roadside lidar data,’’ Sensors, vol. 20, no. 8, p. 2342, 2020.
  • [5] Y. Zhao, J. Zheng,W.Wong, X.Wang, Y. Meng, and H. X. Liu, ‘‘Various methods for queue length and traffic volume estimation using probe vehicle trajectories,’’ Transportation Research Part C: Emerging Technologies, vol. 107, pp. 70–91, 2019.
  • [6] G. Comert, T. Amdeberhan, N. Begashaw, N. G. Medhin, and M. Chowdhury, ‘‘Simple analytical models for estimating the queue lengths from probe vehicles at traffic signals: A combinatorial approach for nonparametric models,’’ Expert Systems with Applications, vol. 252, p. 124076, 2024.
  • [7] Q. Zhou, R. Mohammadi, W. Zhao, K. Zhang, L. Zhang, Y. Wang, C. Roncoli, and S. Hu, ‘‘Queue profile identification at signalized intersections with highresolution data from drones,’’ in 2021 7th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS), pp. 1–6, IEEE, 2021.
  • [8] S. Jayatilleke, V. Wickramasinghe, and N. Amarasingha, ‘‘Introduction of a simple estimation method for lane-based queue lengths with lane-changing movements,’’ Journal of The Institution of Engineers (India): Series A, vol. 104, no. 1, pp. 143–153, 2023.
  • [9] M. Y. Arafat, A. S. M. Khairuddin, U. Khairuddin, and R. Paramesran, ‘‘Systematic review on vehicular licence plate recognition framework in intelligent transport systems,’’ IET intelligent transport systems, vol. 13, no. 5, pp. 745–755, 2019.
  • [10] D. Liu, C. An, M. Yasir, J. Lu, and J. Xia, ‘‘A machine learning based method for real-time queue length estimation using license plate recognition and gps trajectory data,’’ KSCE Journal of Civil Engineering, vol. 26, no. 5, pp. 2408–2419, 2022.
  • [11] X. Zhan, R. Li, and S. V. Ukkusuri, ‘‘Lane-based real-time queue length estimation using license plate recognition data,’’ Transportation Research Part C: Emerging Technologies, vol. 57, pp. 85–102, 2015.
  • [12] P. Pudasaini, A. Karimpour, and Y.-J. Wu, ‘‘Real-time queue length estimation for signalized intersections using single-channel advance detector data,’’ Transportation research record, vol. 2677, no. 7, pp. 144–156, 2023.
  • [13] W. Al Okaishi, A. Zaarane, I. Slimani, I. Atouf, and M. Benrabh, ‘‘A vehicular queue length measurement system in real-time based on ssd network,’’ Transport and Telecommunication, vol. 22, no. 1, pp. 29–38, 2021.
  • [14] Ultralytics, ‘‘Yolov5.’’ https://github.com/ultralytics/yolov5?tab=readme-ov-file. (Accessed: 2023-02-10).
  • [15] A. Battal, Y. Avci, and A. Tuncer, ‘‘Vehicle detection and counting in traffic videos using deep learning,’’ ICENTE’23, p. 272, 2023.
  • [16] M. Boneh, ‘‘Vehicle detection.’’ https://github.com/MaryamBoneh/VehicleDetection/tree/main/Dataset. (Accessed: 2023-02-12).
  • [17] L. Soetanto, ‘‘Vehicle detection.’’ https://www.kaggle.com/datasets/lyensoetanto/vehicle-images-dataset. (Accessed: 2023-02-21).
  • [18] M. TÁRNOK, ‘‘5 vehichles for classification.’’ https://www.kaggle.com/datasets/mrtontrnok/5-vehichles-for-multicategory-classification. (Accessed: 2024-06- 06).
  • [19] A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, ‘‘Simple online and realtime tracking,’’ in 2016 IEEE international conference on image processing (ICIP), pp. 3464–3468, Ieee, 2016.
  • [20] P. K. Thadagoppula and V. Upadhyaya, ‘‘Speed detection using image processing,’’ in 2016 International Conference on Computer, Control, Informatics and its Applications (IC3INA), pp. 11–16, IEEE, 2016.
  • [21] C.-J. Lin, S.-Y. Jeng, and H.-W. Lioa, ‘‘A real-time vehicle counting, speed estimation, and classification system based on virtual detection zone and yolo,’’ Mathematical Problems in Engineering, vol. 2021, no. 1, p. 1577614, 2021.
  • [22] A. M. Santos, C. J. Bastos-Filho, A. M. Maciel, and E. Lima, ‘‘Counting vehicle with high-precision in brazilian roads using yolov3 and deep sort,’’ in 2020 33rd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), pp. 69–76, IEEE, 2020.
  • [23] Y. Zhao, X. Zhou, X. Xu, Z. Jiang, F. Cheng, J. Tang, and Y. Shen, ‘‘A novel vehicle tracking id switches algorithm for driving recording sensors,’’ Sensors, vol. 20, no. 13, p. 3638, 2020.
Toplam 23 adet kaynakça vardır.

Ayrıntılar

Birincil Dil İngilizce
Konular Mühendislik Uygulaması
Bölüm Araştırma Makaleleri
Yazarlar

Ahsen Battal 0000-0002-4824-5889

Yunus Emre Avcı 0000-0003-3921-7162

Adem Tuncer 0000-0001-7305-1886

Proje Numarası 2023/AP/0002
Yayımlanma Tarihi 30 Eylül 2025
Gönderilme Tarihi 1 Ağustos 2025
Kabul Tarihi 16 Eylül 2025
Yayımlandığı Sayı Yıl 2025 Cilt: 12 Sayı: 3

Kaynak Göster

IEEE A. Battal, Y. E. Avcı, ve A. Tuncer, “Real-Time Detection of Vehicle Queue States in Urban Traffic Using Deep Learning”, ECJSE, c. 12, sy. 3, ss. 356–364, 2025, doi: 10.31202/ecjse.1755333.