Research on a Deep Learning-Based Lane Line Recognition System

DOI: https://doi.org/10.62517/jbdc.202601316

Author(s)

Yirui Liu, Aimei Xiao

Affiliation(s)

School of Computer Science and Technology, Shandong University of Technology, Zibo, Shandong, China

Abstract

The identification of lane lines and the detection of road objects are important parts of environmental perception for intelligent driving. The current methods for lane detection are usually influenced by complicated backgrounds, changes in lighting and shadows, as well as the quality of the video, and a single detection function is not enough to fully understand the road environment. To solve these problems, a deep learning-based lane line recognition system has been developed and used in this paper. The system is implemented using Python, OpenCV, PyQt5, PyTorch and YOLOv5, including video pre-processing, lane line detection, road object recognition, video clipping and saving, uploading the video to Android side and measuring distance with binocular LiDAR cameras. The lane detection module identifies the left and right lane lines by using grayscale transformation, Canny edge detection, region of interest masking, Hough line detection and slope filtering. The road object recognition module employs YOLOv5 for vehicle and object detection and performs non-maximum suppression to improve the detection results of the bounding boxes. Experimental results indicate that the object-box selection accuracy is over 95% in three test videos, and the lane line recognition accuracy is also good. The system can stably detect and visualize road-scene information from video inputs, providing a practical reference for lightweight assisted-driving perception systems.

Keywords

Deep Learning; Lane Line Recognition; Yolov5; Opencv; Intelligent Driving; Object Detection

References

[1] Assidiq A A M, Khalifa O O, Islam M R, et al. Real time lane detection for autonomous vehicles//2008 International Conference on Computer and Communication Engineering. IEEE, 2008: 82-88. [2] Redmon J, Farhadi A. YOLO9000: better, faster, stronger//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 7263-7271. [3] He X, Guo H, Zhu K, et al. Monocular lane detection based on deep learning: a survey. arXiv:2411.16316, 2024. [4] Swain S, Tripathy A K. Real-time lane detection for autonomous vehicles using YOLOV5 Segmentation Model. International Journal of Sustainable Engineering, 2024, 17(1): 718-728. [5] Maddiralla V, Subramanian S. Effective lane detection on complex roads with convolutional attention mechanism in autonomous vehicles Scientific Reports, 2024, 14: 19193. [6] Yordanov D, Chakraborty A, Hasan M M, Cirstea S. A framework for optimizing deep learning-based lane detection and steering for autonomous driving. Sensors, 2024, 24(24): 8099. [7] Feilong Q, Khan N A, Jhanjhi N Z, Ashfaq F, Hendrawati T D. Improved YOLOv5 lane line real time segmentation system integrating Seg Head Network. Engineering Proceedings, 2025, 107(1): 49. [8] Chen L C, Papandreou G, Kokkinos I, et al. Deeplab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 40(4): 834-848. [9] Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation//Medical Image Computing and Computer-Assisted Intervention. Springer, 2015: 234-241. [10]Zou Q, Jiang H, Dai Q, et al. Robust lane detection from continuous driving scenes using deep neural networks. IEEE Transactions on Vehicular Technology, 2019, 69(1): 41-54. [11]Zhang Y, Lu Z, Zhang X, et al. Deep learning in lane marking detection: a survey. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(7): 5976-5992. [12]Tang J, Li S, Liu P. A review of lane detection methods based on deep learning. Pattern Recognition, 2021, 111: 107623. [13]Wang Z, Zhan J, Duan C, et al. A review of vehicle detection techniques for intelligent vehicles. IEEE Transactions on Neural Networks and Learning Systems, 2022, 34(8): 3811-3831. [14]Bojarski M, Del Testa D, Dworakowski D, et al. End to end learning for self-driving cars. arXiv preprint arXiv:1604.07316, 2016. [15]Rahman Z, Morris B T. LVLane: deep learning for lane detection and classification in challenging conditions//2023 IEEE 26th International Conference on Intelligent Transportation Systems. IEEE, 2023: 3901-3907. [16]Li S Z. Design and implementation of a lane line detection system based on deep learning. Harbin Institute of Technology, 2016. [17]Wang J Q. Design and implementation of a lane line detection system based on deep learning. Southwest University, 2022. [18]Yang H. Design and implementation of a highway road-condition detection system. University of Electronic Science and Technology of China, 2023.