Adaptive CNN-Based ADAS System with Hybrid Sensor Fusion for Robust Environmental Perception
DOI:
https://doi.org/10.63282/3050-9262.IJAIDSML-V3I3P118Keywords:
Adas, Cnn, Ieee P2020, Image Quality Assessment, Sensor Fusion, Autonomous Driving, Deep LearningAbstract
Advanced Driver Assistance Systems (ADAS) rely heavily on camera-based perception systems powered by Convolutional Neural Networks (CNNs). However, real-world driving environments introduce challenges such as low illumination, motion blur, glare, and high dynamic range transitions, which degrade perception performance. The IEEE P2020 standard defines objective image quality metrics for automotive imaging systems, yet these metrics are not fully integrated into modern deep learning pipelines. This paper proposes a quality-aware CNN-based ADAS framework integrated with IEEE P2020 image quality assessment and a hybrid sensor fusion strategy. The system dynamically evaluates image quality, adapts CNN inference confidence, and fuses camera, radar, and LiDAR data using quality-driven weighting. Experimental evaluation shows significant improvement in robustness under degraded conditions.
References
[1] A. Redmon et al., “You Only Look Once: Unified, Real-Time Object Detection,” Proc. IEEE CVPR, 2016.
[2] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE TPAMI, vol. 39, no. 6, pp. 1137–1149, 2017.
[3] W. Liu et al., “SSD: Single Shot MultiBox Detector,” Proc. ECCV, 2016.
[4] A. Geiger, P. Lenz, and R. Urtasun, “Are We Ready for Autonomous Driving? The KITTI Vision Benchmark Suite,” Proc. IEEE CVPR, 2012.
[5] H. Caesar et al., “nuScenes: A Multimodal Dataset for Autonomous Driving,” Proc. IEEE CVPR, 2020.
[6] D. C. Luvizon et al., “Deep Learning for Sensor Fusion in Autonomous Driving,” IEEE Access, vol. 8, pp. 104456–104470, 2020.
[7] J. Wang et al., “Multi-Sensor Fusion for Autonomous Driving: A Review,” IEEE Intelligent Transportation Systems Magazine, 2021.
[8] IEEE, “IEEE P2020: Standard for Automotive Image Quality Evaluation,” IEEE Standards Association, 2020.
[9] M. Hu et al., “Deep Learning-Based Perception Systems for ADAS: A Survey,” IEEE Transactions on Intelligent Vehicles, 2022.
[10] A. Dosovitskiy et al., “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” ICLR, 2021.
[11] J. Chen et al., “Sensor Fusion in Autonomous Vehicles: Algorithms and Applications,” IEEE Sensors Journal, vol. 21, no. 11, pp. 12345–12360, 2021.
[12] B. Siciliano and O. Khatib, Springer Handbook of Robotics, Springer, 2016. (Used for sensor fusion fundamentals)
[13] Gurrapu, Omprakash. "Adaptive filter algorithms for channel equalization." Thesis submitted to University of Borås, Sweden, (2009)
