Improved YOLOv11-Based Object Detection Algorithm for Autonomous Driving in Low-Light Conditions

XiYe Luo

doi:10.63313/AERpc.9074

Authors

XiYe Luo Tianjin University of Technology and Education, Tianjin, 300222, China Author

DOI:

https://doi.org/10.63313/AERpc.9074

Keywords:

low-light object detection, YOLOv11, autonomous driving, deep learning

Abstract

To address the severe challenges of blurred object features and strong background interference in low-light nighttime environments, which often lead to false detections and missed detections in autonomous driving object detection, this paper proposes an improved YOLOv11-based object detection algorithm for autonomous driving in low-light conditions (IEWS-YOLO). First, an inverted residual attention mechanism called the iEMA module is designed and inserted into the backbone network to enhance detection performance in low-light environments. Second, the standard convolutions in the backbone network are replaced with wavelet transform convolution layers (WTConv), improving global feature representation efficiency and noise suppression capability. Finally, the traditional CIoU loss function is replaced with the ShapeIoU loss function to optimize bounding box regression accuracy, making predicted box locations more precise and enhancing object localization. Experimental results show that the IEWS-YOLO model achieves significant improvements in various performance metrics compared to the traditional YOLOv11n model. On the ExDark standard low-light dataset, IEWS-YOLO improves [email protected] by 9.6% and [email protected]:0.95 by 6.7% compared to YOLOv11n. In conclusion, the IEWS-YOLO model can perform object detection for autonomous driving in low-light environments more accurately.

References

[1] Zhang W ,Xu H ,Zhu X , et al. RFSC-net: Re-parameterization forward semantic compensation network in low-light environments[J]. Image and Vision Computing,2024,151105271-105271.DOI:10.1016/J.IMAVIS.2024.105271.

[2] He, Lh., Zhou, Yz., Liu, L. et al. Research on object detection and recognition in remote sensing images based on YOLOv11. Sci Rep 15, 14032 (2025).

[3] Lin, J., Wang, P., Ruan, Y. et al. YOLO11-WLBS: an efficient model for pavement defect detection. Sci Rep (2026). https://doi.org/10.1038/s41598-026-35743-8

[4] Ren, B., Xu, Z., Zhao, J. et al. Complex dark environment-oriented object detection method based on YOLO-AS. Sci Rep 15, 21873 (2025). https://doi.org/10.1038/s41598-025-07348-0

[5] Yang, S. et al. LightingNet: An integrated learning method for low-light image enhancement. IEEE Trans. Comput. Imaging 9, 29–42 (2023).

[6] Li, X., Pu, X., Ling, W. et al. YOLO-SAM an end-to-end framework for efficient real time object detection and segmentation. Sci Rep 15, 40854 (2025). https://doi.org/10.1038/s41598-025-24576-6

[7] Han Y, Qi K, Zheng J, et al. Lightweight Cattle Facial Recognition Method Based on Improved YOLOv11. Smart Agriculture, 2025, 7(3): 173-184. https://doi.org/10.12133/j.smartag.SA202502010

[8] Xu, X. et al. Exploring image enhancement for salient object detection in low light images. ACM Trans. Multimed. Comput. Commun. Appl. 17, 1–19 (2021).

[9] Wang J ,Yang P ,Liu Y , et al. Research on Improved YOLOv5 for Low-Light Environment Object Detection[J]. Electronics,2023,12(14):DOI:10.3390/ELECTRONICS12143089.

[10] Wang Yin, Wang Lide, Qiu Ji. Real-time Enhancement Algorithm for Orbital Dark Light Environment Based on DenseNet Structure[J]. Journal of Southwest Jiaotong University, 2022, 57(6): 1349-1357. doi: 10.3969/j.issn.0258-2724.20210199

[11] Xue R ,Duan J ,Du Z .MPE-DETR: A multiscale pyramid enhancement network for object detection in low-light images[J].Image and Vision Computing,2024,150105202-105202.DOI:10.1016/J.IMAVIS.2024.105202.