Köpeczi-Bócz, Ákos Tamás and Mi, Tian and Orosz, Gábor and Takács, Dénes (2024) YOLOgraphy: Image processing based vehicle position recognition. LECTURE NOTES IN MECHANICAL ENGINEERING. pp. 392-398. ISSN 2195-4356
|
Text
AVEC_2024_Akos.pdf - Published Version Download (8MB) | Preview |
Abstract
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.
| Item Type: | Article |
|---|---|
| Additional Information: | Correspondence Address: Köpeczi-Bócz, Á.T.; Department of Applied Mechanics, Hungary; email: kopeczi@mm.bme.hu Funding details: Budapesti Műszaki és Gazdaságtudományi Egyetem, BME Funding details: Magyar Tudományos Akadémia, MTA Funding details: Nemzeti Kutatási, Fejlesztési és Innovaciós Alap, NKFIA Funding details: University of Michigan, U-M Funding details: Nemzeti Kutatási Fejlesztési és Innovációs Hivatal, NKFIH, NKFI-146201 Funding details: U.S. Department of Transportation, DOT, 69A3552348305 Funding text 1: The research reported in this paper was supported by the J\\u00E1nos Bolyai Research Scholarship of the Hungarian Academy of Sciences, the National Research, Development and Innovation Office under grant no. NKFI-146201, and the University of Michigan\\u2019s Center for Connected and Automated Transportation through the US DOT grant 69A3552348305. The project supported by the Doctoral Excellence Fellowship Programme (DCEP) is funded by the National Research Development and Innovation Fund of the Ministry of Culture and Innovation and the Budapest University of Technology and Economics, under a grant agreement with the National Research, Development and Innovation Office. The research is partly supported by the Foundation for Mechanical Engineering Education. |
| Uncontrolled Keywords: | Image processing, Vehicle dynamics, Machine learning |
| Subjects: | T Technology / alkalmazott, műszaki tudományok > TA Engineering (General). Civil engineering (General) / általános mérnöki tudományok T Technology / alkalmazott, műszaki tudományok > TJ Mechanical engineering and machinery / gépészmérnöki tudományok |
| SWORD Depositor: | MTMT SWORD |
| Depositing User: | MTMT SWORD |
| Date Deposited: | 19 Sep 2025 10:18 |
| Last Modified: | 19 Sep 2025 10:22 |
| URI: | https://real.mtak.hu/id/eprint/224618 |
Actions (login required)
 |
Edit Item |
