In response to the inadequacies of existing traffic surveillance systems, this study presents a cutting-edge approach employing YOLOv8 for detection and ByteTrack for tracking, enhancing real-time traffic monitoring capabilities. Our system, designed for minimal computational demands, excels in the rapid identification and tracking of road targets. By specifically tailoring the YOLOv8n detection model for vehicular recognition and combining it with ByteTrack’s efficient tracking, we achieve precise vehicle classification, detection, and tracking. This methodology allows for accurate, real-time calculations of instantaneous and cumulative traffic flows by monitoring unique vehicle IDs over time. The integration of YOLOv8n and ByteTrack combines the former’s accuracy in detection with the latter’s speed in tracking, alongside their mutual benefits in swift processing, ensuring a high-performance surveillance system that significantly enhances traffic management and planning. Our contribution marks a significant leap towards smarter, sustainable urban traffic monitoring systems.
Enhancing Traffic Surveillance Through YOLOv8 and ByteTrack: A Novel Approach to Real-Time Detection, Classification, and Tracking
Lect. Notes Electrical Eng.
International Conference on Machine Learning, Cloud Computing and Intelligent Mining ; 2024 ; Shennongjia, China August 08, 2024 - August 11, 2024
Proceedings of the 3rd International Conference on Machine Learning, Cloud Computing and Intelligent Mining (MLCCIM2024) ; Chapter : 23 ; 271-279
2025-03-22
9 pages
Article/Chapter (Book)
Electronic Resource
English