Self-Supervised Learning for Robust Lane Geometry Estimation under Adverse Weather
Keywords:
Self-Supervised Learning, Lane Geometry Estimation, Adverse Weather Robustness, Autonomous Driving, Advanced Driver Assistance Systems (ADAS), Temporal Consistency, Geometric Constraints, Monocular Vision, Attention Mechanisms, Uncertainty Modeling.Abstract
Robust Lane geometry estimation is a critical component of advanced driver assistance systems (ADAS) and autonomous driving, yet its reliability significantly degrades under adverse weather conditions such as rain, fog, snow, and low illumination. This paper presents a self-supervised learning framework for resilient lane geometry estimation without reliance on extensive pixel-level annotations. The proposed approach leverages temporal consistency, geometric constraints, and cross-view photometric alignment to generate supervisory signals directly from raw monocular video streams. By integrating weather-aware feature normalization and uncertainty-guided refinement modules, the model enhances robustness against visibility degradation and sensor noise. A hybrid encoder–decoder architecture with attention-based spatial aggregation captures both local lane markings and global structural priors of road topology. Additionally, a geometry-consistency loss enforces structural coherence across sequential frames, mitigating spurious predictions caused by occlusions and dynamic artifacts. Extensive evaluations on benchmark driving datasets augmented with synthetic and real-world adverse weather conditions demonstrate superior generalization compared to fully supervised baselines trained on limited labeled data. The proposed method achieves improved lane curvature estimation accuracy and boundary continuity while reducing annotation dependency. These results underscore the potential of self-supervised paradigms in enabling scalable, weather-resilient perception systems for next-generation autonomous vehicles.
Downloads
References
. N D. Kumar and N. Muhammad, “Object Detection in Adverse Weather for Autonomous Driving through Data Merging and YOLOv8,” Sensors, vol. 23, no. 20, 2023.
. Maddiralla and S. Subramanian, “Effective Lane Detection on Complex Roads with Convolutional Attention Mechanism in Autonomous Vehicles,” Scientific Reports, vol. 14, 2024.
. J. Lindberg, “Self-Supervised Deep Learning for Autonomous Vehicle Perception under Adverse Weather Conditions,” Future AI & Intelligent Transportation Systems, 2025.
. K. Saunders, G. Vogiatzis, and L. Manso, “Self-Supervised Monocular Depth Estimation: Let’s Talk About the Weather,” arXiv preprint, 2023.
. J. Wang et al., “WeatherDepth: Curriculum Contrastive Learning for Self-Supervised Depth Estimation under Adverse Weather Conditions,” arXiv preprint, 2023.
. L. Deng et al., “Semi-Supervised Lane Detection for Continuous Traffic Scenes,” Traffic Injury Prevention, vol. 24, no. 6, 2023.
. “End-to-End Vehicle Multi-Task Perception in Adverse Weather,” Robotics and Autonomous Systems, 2025.
. Maddiralla and S. Subramanian, “Attention-Based Lane Detection in Complex Environments,” Scientific Reports, 2024.
. F. Garcea et al., “Self-Supervised and Semi-Supervised Learning for Road Condition Estimation,” Scientific Reports, vol. 12, 2022.
. P. P. Anoop and R. Deivanathan, “Real-Time Road Scene Classification and Enhancement under Adverse Weather,” Scientific Reports, 2025.
. X. Pan, J. Shi, P. Luo, X. Wang, and X. Tang, “Spatial As Deep: Spatial CNN for Traffic Scene Understanding,” in Proc. AAAI Conf. Artificial Intelligence, 2018, pp. 7276–7283.
. C. Sakaridis, D. Dai, and L. Van Gool, “Semantic Foggy Scene Understanding with Synthetic Data,” Int. J. Computer Vision, vol. 126, no. 9, pp. 973–992, 2018.
. T. Chen, S. Kornblith, M. Norouzi, and G. Hinton, “A Simple Framework for Contrastive Learning of Visual Representations (SimCLR),” in Proc. Int. Conf. Machine Learning (ICML), 2020, pp. 1597–1607.
. K. He, H. Fan, Y. Wu, S. Xie, and R. Girshick, “Momentum Contrast for Unsupervised Visual Representation Learning (MoCo),” in Proc. IEEE/CVF Conf. Computer Vision and Pattern Recognition (CVPR), 2020, pp. 9729–9738.
. T. Zhou, M. Brown, N. Snavely, and D. G. Lowe, “Unsupervised Learning of Depth and Ego-Motion from Video,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2017, pp. 1851–1858.
. Y. Zhang, Q. Yang, and X. Li, “Uncertainty-Guided Self-Supervised Learning for Robust Visual Perception under Adverse Weather,” IEEE Trans. Intelligent Transportation Systems, 2023.
. S. Samraj, “Impacts of model based design in avionics software,” International Journal of Innovative Engineering and Management Research, vol. 10, no. 12, 2021, doi: 10.48047/IJIEMR/V10/ISSUE12/50.
. Hari Prasad Bhupathi,Srikiran Chinta, Dr Vijayalaxmi Biradar, Dr Sanjay Kumar Suman. (2023). Artificial Intelligence and Machine Learning Solutions for Efficient Battery Management and Balancing . Journal of Computational Analysis and Applications (JoCAAA), 31(4), 1545–1554C.
. S. Samraj, "Avionics systems integration using avionics full duplex swithched ethernet," 2007 IEEE/AIAA 26th Digital Avionics Systems Conference, Dallas, TX, USA, 2007, pp. 2.E.4-1-2.E.4-1, doi: 10.1109/DASC.2007.4391867
. S. Samraj, “Impacts of model based design in avionics software,” International Journal of Innovative Engineering and Management Research, vol. 10, no. 12, 2021, doi: 10.48047/IJIEMR/V10/ISSUE12/50.
. Selvadhas Samraj, Saicharan Allenki and Merlin M, “Automated Fault Injection Framework for Functional Safety Validation of Autonomous Driver Assistance System”, IJIEE, vol. 13, no. 4, pp. 148–154, Dec. 2023, doi: 10.48047/x0vqt067.
. S. Samraj, S. Allenki, and M. Merlin, “A unified safety model for predictive decision making in level 2+ and level 3 autonomous driving systems,” Res Militaris, vol. 13, no. 4, Dec. 2023.
. SelvadhasSamraj, Saicharan Allenki, Merlin M,. (2024). Predictive Battery Management for Electric Vehicles Using Adaptive State Estimation Models. Journal of Computational Analysis and Applications (JoCAAA), 33(08), 8660–8668. Retrieved from https://www.eudoxuspress.com/index.php/pub/article/view/5413
. S. Sabaria, Sindhu Ravindran, Bhavani R., C. Laxmikanth Reddy, CH. M. H. Saibaba, Saggurthi Rajesh, L. Bhagyalakshmi (2024). Next-Generation Spatial Data Management Leveraging Spatial Databases and Blockchain in Cloud Data Architectures. International Journal of Intelligent Systems and Applications in Engineering, 12(21s), 1291–1300.
. Chandrasekaran, R. ., B. V., S. K. ., Loganathan, B. ., Suman, S. K. ., & Bhagyalakshmi. (2023). Glaucoma Detection with Improved Deep Learning Model Trained with Optimal Features: An Improved Meta-Heuristic Model. International Journal of Intelligent Systems and Applications in Engineering, 11(6s), 532–547.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
