A methodology is developed to extract vehicle kinematic information from roadside cameras at an intersection using deep learning. The ground truth data of top view bounding boxes are collected with the help of unmanned aerial vehicles (UAVs). These top view bounding boxes containing vehicle position, size, and orientation information, are converted to the roadside view bounding boxes using homography transformation. The ground truth data and the roadside view images are used to train a modified YOLOv5 neural network, and thus, to learn the homography transformation matrix. The output of the neural network is the vehicle kinematic information, and it can be visualized in both the top view and the roadside view. In our algorithm, the top view images are only used in training, and once the neural network is trained, only the roadside cameras are needed to extract the kinematic information.
YOLOgraphy: Image Processing Based Vehicle Position Recognition
Lect.Notes Mechanical Engineering
Advanced Vehicle Control Symposium ; 2024 ; Milan, Italy September 01, 2024 - September 05, 2024
2024-10-04
7 pages
Article/Chapter (Book)
Electronic Resource
English
Image Processing based Vehicle Number Plate Recognition
IEEE | 2022
|OWN VEHICLE POSITION RECOGNITION DEVICE AND OWN VEHICLE POSITION RECOGNITION METHOD
European Patent Office | 2016
|POSITION RECOGNITION METHOD AND POSITION RECOGNITION SYSTEM FOR VEHICLE
European Patent Office | 2024
|POSITION RECOGNITION METHOD AND POSITION RECOGNITION SYSTEM FOR VEHICLE
European Patent Office | 2022
|