Application of Vehicle detection and tracking model: Estimation of traffic flow variable based on Moving Observer Method
Article Sidebar
Main Article Content
Abstract
Traffic surveys are the foundation for analysis, design, planning, assessment and management of traffic and transportation. The Moving Observer Method (MOM) is widely used for macroscopic traffic flow variable survey which is easy to apply and, cost and time efficient. This study applied artificial Intelligence (AI) techniques for traffic surveys, using YOLOv7 architecture for vehicle detection, StrongSORT architecture for vehicle tracking, Canny Edge and Hough Transform for lane line detection to classify vehicle type and movement for estimating traffic flow variables based on MOM. The experiment was implemented with datasets from 5intercity routes in Thailand, single carriageway road with
2 lanes, length 2 to 5 km. The results indicate that the performance of vehicle type and movement classification (number of opposing vehicles, vehicles overtaking the test car, vehicles passed by the test car) was F1-score of 93.25, 94.79, and 64.62% respectively. In summary, the performance of traffic flow variable estimation based on MOM (flow rate, mean speed and density) compared to the actual data was mean absolute percentage error of 2.36, 0.73 and 2.86% respectively and highest absolute percentage error was 7.88, 5.80 and 7.65% respectively.
Article Details
References
Alhomaidat F, Ardekani S. A Statistical Comparison of Traffic Measurements from the Moving versus Stationary Observer Methods. Journal of Transportation Technologies. 2015;5: 204-13.
Mulligan A-M, Nicholson A. Uncertainty in traffic flow estimation using the moving-observer method. IPENZ Transportation Group Conference Papers; 2002.
Wardrop JG, Charlesworth G. A Method of Estimating Speed and Flow of Traffic from a Moving Vehicle. Proceedings of the Institution of Civil Engineers. 1954;3(1): 158-71.
Greenshields B, Bibbins J, Channing W, Miller H. A study of traffic capacity. Highway research board proceedings; 1935. p. 448-77.
Xiao Y, Tian Z, Yu J, Zhang Y, Liu S, Du S, et al. A review of object detection based on deep learning. Multimedia Tools and Applications. 2020;79(33): 23729-91.
Ren S, He K, Girshick R, Sun J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2017;39(6): 1137-49.
Wang C-Y, Bochkovskiy A, Liao H-YM. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. arXiv preprint arXiv:220702696. 2022.
Yang F, Zhang X, Liu B. Video object tracking based on YOLOv7 and DeepSORT. arXiv preprint arXiv:2207 02696. 2022.
Wojke N, Bewley A, Paulus D. Simple online and realtime tracking with a deep association metric. 2017 IEEE International Conference on Image Processing (ICIP); 2017. p. 3645-9.
Du Y, Song Y, Yang B, Zhao Y. StrongSORT: Make DeepSORT Great Again. arXiv e-prints. 2022: arXiv:2202.13514.
Zhang Y, Wang C, Wang X, Zeng W, Liu W. Fairmot: On the fairness of detection and re-identification in multiple object tracking. International Journal of Computer Vision. 2021;129(11): 3069-87.
Kumar A, Simon P. Review of Lane Detection and Tracking Algorithms in Advanced Driver Assistance System. International Journal of Computer Science and Information Technology. 2015;7: 65-78.
Priyadharshini P, Niketha P, Lakshmi KS, Sharmila S, Divya R. Advances in Vision based Lane Detection Algorithm Based on Reliable Lane Markings. 2019 5th International Conference on Advanced Computing & Communication Systems 2019. p. 880-5.
Guerrieri M, Parla G, Mauro R. Traffic Flow Variables Estimation: An Automated Procedure Based on Moving Observer Method. Potential Application for Autonomous Vehicles. Transport and Telecommunication Journal. 2019;20: 205-14.
Guerrieri M, Parla G. Deep Learning and YOLOv3 Systems for Automatic Traffic Data Measurement by Moving Car Observer Technique. Infrastructures. 2021;6(9): 134.
Kraisingsom S, Phawiakkharakun N. Traffic Congestion Index and Density Analysis Report in 2021. Bureau of Highway Safety, Department of Highways, Thailand; 2021.
Al-Bahr T, Puan OC, Hassan SA, Azhari SF, Al-Sabaeei AM, Mohammed HA. Statistical Evaluation of Moving Observer Method Accuracy for Measuring Traffic Flow Variables on Urban Roads. 2021 International Congress of Advanced Technology and Engineering (ICOTEN); 2021. p. 1-6.
Mortimer WJ. Moving vehicle method of estimating traffic volumes and speeds. Highway Research Board Bulletin. 1957;156: 14-26.
Al-Bahr TM, Puan OC, Hassan SA, Idham MK, Ismail CR. Parameters affecting fluctuation in traffic stream on urban roads. IOP Conference Series: Materials Science and Engineering. 2019;527(1): 012067.
Broström M. Real-time multi-camera multi-object tracker using YOLOv7 and StrongSORT with OSNet. Available from: https://github.com/mikel-brostrom/Yolov7_StrongSORT_OSNet. [Accessed 21th August 2020].
Vykari N. Understanding Hough Transform With A Lane Detection Model. Available from: https://blog.paperspace.com/understanding-hough-transform-lane-detection/. [Accessed 13th February 2022].