A Method for Building Vehicle Trajectory Data Sets Based on Drone Videos

2023-01-0714

04/11/2023

Event
WCX SAE World Congress Experience
Authors Abstract
Content
The research and development of data-driven highly automated driving system components such as trajectory prediction, motion planning, driving test scenario generation, and safety validation all require large amounts of naturalistic vehicle trajectory data. Therefore, a variety of data collection methods have emerged to meet the growing demand. Among these, camera-equipped drones are gaining more and more attention because of their obvious advantages. Specifically, compared to others, drones have a wider field of bird's eye view, which is less likely to be blocked, and they could collect more complete and natural vehicle trajectory data. Besides, they are not easily observed by traffic participants and ensure that the human driver behavior data collected is realistic and natural. In this paper, we present a complete vehicle trajectory data extraction framework based on aerial videos. It consists of three parts: 1) objects detection, 2) data association, and 3) data cleaning. In particular, considering that the hovering drone can be approximated as a fixed camera, we propose an improved object detection algorithm based on classical image processing algorithms. It overcomes the shake effects of drone-based aerial videos and can be directly applied to the automatic detection of moving vehicles without manual annotation data. The output of the algorithm is the vehicle rotated bounding box information with high accuracy, including vehicle center position, vehicle heading, and vehicle dimension. In addition, the improved detection algorithm can be used for vehicle object automatic annotation.
Meta TagsDetails
DOI
https://doi.org/10.4271/2023-01-0714
Pages
7
Citation
Wang, Z., Yu, Z., Tian, W., Xiong, L. et al., "A Method for Building Vehicle Trajectory Data Sets Based on Drone Videos," SAE Technical Paper 2023-01-0714, 2023, https://doi.org/10.4271/2023-01-0714.
Additional Details
Publisher
Published
Apr 11, 2023
Product Code
2023-01-0714
Content Type
Technical Paper
Language
English