Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jul 7;23(13):6238.
doi: 10.3390/s23136238.

Research on Road Scene Understanding of Autonomous Vehicles Based on Multi-Task Learning

Jinghua Guo  1 , Jingyao Wang  2 , Huinian Wang  1 , Baoping Xiao  1 , Zhifei He  1 , Lubin Li  1
Affiliations

Research on Road Scene Understanding of Autonomous Vehicles Based on Multi-Task Learning

Jinghua Guo et al. Sensors (Basel). .

Abstract

Road scene understanding is crucial to the safe driving of autonomous vehicles. Comprehensive road scene understanding requires a visual perception system to deal with a large number of tasks at the same time, which needs a perception model with a small size, fast speed, and high accuracy. As multi-task learning has evident advantages in performance and computational resources, in this paper, a multi-task model YOLO-Object, Drivable Area, and Lane Line Detection (YOLO-ODL) based on hard parameter sharing is proposed to realize joint and efficient detection of traffic objects, drivable areas, and lane lines. In order to balance tasks of YOLO-ODL, a weight balancing strategy is introduced so that the weight parameters of the model can be automatically adjusted during training, and a Mosaic migration optimization scheme is adopted to improve the evaluation indicators of the model. Our YOLO-ODL model performs well on the challenging BDD100K dataset, achieving the state of the art in terms of accuracy and computational efficiency.

Keywords: autonomous vehicles; drivable area detection; lane line detection; multi-task learning; traffic object detection; visual perception.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Our road detection.
Figure 2
Figure 2
Structure of the proposed YOLO-ODL model.
Figure 3
Figure 3
Data augmentation.
Figure 4
Figure 4
Migration optimization.
Figure 5
Figure 5
Detection results of YOLO-ODL in multi-weather road conditions.
Figure 6
Figure 6
Detection results of YOLO-ODL in multi-scenario road conditions.
Figure 7
Figure 7
Comparison of YOLO-ODL and YOLOP object detection results. The red box in the figure is the false detection box, and the yellow box is the missed detection box.
Figure 8
Figure 8
Comparison of YOLO-ODL and YOLOP drivable area detection results. The red circle in the figure is the false detection area, and the yellow circle is the missed detection area.
Figure 9
Figure 9
Comparison of YOLO-ODL and YOLOP lane line detection results. The yellow circle is the missed detection area.
See this image and copyright information in PMC

References

    1. Qian Y., Dolan J.M., Yang M. DLT-Net: Joint detection of drivable areas, lane lines, and traffic objects. IEEE Trans. Intell. Transp. Syst. (IVS) 2019;21:4670–4679. doi: 10.1109/TITS.2019.2943777. - DOI
    1. Teichmann M., Weber M., Zollner M., Cipolla R., Urtasun R. MultiNet: Real-time joint semantic reasoning for autonomous driving; Proceedings of the 2018 IEEE Intelligent Vehicles Symposium (IV); Changshu, China. 26–30 June 2018; pp. 1013–1020.
    1. Sun Z., Bebis G., Miller R. On-road vehicle detection: A review. IEEE Trans. Pattern Anal. Mach. Intell. 2006;28:694–711. - PubMed
    1. Owais M. Traffic sensor location problem: Three decades of research. Expert Syst. Appl. 2022;208:118134. doi: 10.1016/j.eswa.2022.118134. - DOI
    1. Bhaggiaraj S., Priyadharsini M., Karuppasamy K., Snegha R. Deep Learning Based Self Driving Cars Using Computer Vision; Proceedings of the 2023 International Conference on Networking and Communications (ICNWC); Chennai, India. 5–6 April 2023; pp. 1–9.