Detecting and tracking multiple target vehicles in traffic videos, locating and extracting their trajectories, and obtaining information such as their speed, are of great importance in practical applications. Therefore, this has been a hot research topic, and using extracted trajectory data for this study addresses the problem of severe traffic conflicts in complex road sections without traffic control, and proposes the use of extracted trajectory data to improve road traffic safety. road sections without traffic control, and proposes a trajectory extraction method based on a deep learning object detection algorithm and a method of The study inputs a road video with mixed traffic of motor vehicles, non-motor vehicles, and pedestrians without traffic control. The study inputs a road video with mixed traffic of motor vehicles, non-motor vehicles, and pedestrians without traffic control, and uses a combination of yolov3 and sort algorithms for trajectory extraction and tracking of By transforming the image coordinate system, camera coordinate system, and world coordinate system, the three-dimensional information of Finally, the position, speed, and other information of each frame of the vehicle are output, and the information is batch- Finally, the position, speed, and other information of each frame of the vehicle are output, and the information is batch-processed and compared with the threshold value of TTC to achieve risk prediction.
Conflict determination of without traffic control based on yolov3 and sort algorithm
2023-08-18
2137203 byte
Conference paper
Electronic Resource
English
Traffic Monitoring System using YOLOv3 Model
IEEE | 2023
|Ship target detection based on YOLOv3 algorithm
IEEE | 2023
|