Abstract
High real-time requirements, complex backgrounds, and small targets are major challenges in traffic sign recognition. Mainstream algorithms still suffer from issues such as complex model structures, redundant parameters, and high computational demands. Therefore, striking a balance between model lightweighting and accuracy has become a new challenge. This paper is based on the single-stage YOLOv5 algorithm framework and presents an optimized design of YOLO-F detection model with fewer parameters, high detection accuracy, the ability to effectively identify small targets, and easy deployment. The detection algorithm achieves an accuracy of 87.5%, a recall rate of 81.7%, and a detection speed of 89FPS. Furthermore, the detection model undergoes LAMP channel pruning to significantly reduce parameters and computational load while maintaining accuracy. Additionally, the pruned model undergoes distillation to enhance accuracy. Experimental results demonstrate that the improved YOLO-FLC model, compared to YOLOv5s, achieves reductions of 80.49%, 39.38%, and 74.22% in model parameters, computational load, and weight, respectively, while increasing mAP to 0.881, up by 2.6%. The FPS reaches 99, effectively balancing real-time performance and accuracy.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Ross, G., et at.: Rich feature hierarchies for accurate object detection and semantic segmentation. IEEE Conference on Computer Vision and Pattern Recognition, June 2014, pp. 580–587 (2014)
Ross, G.: Fast R-CNN. In: IEEE International Conference on Computer Vision (ICCV), Santiago, Chile, 2015, pp. 1440–1448 (2015). [ https://doi.org/10.1109/ICCV.2015.169.]
Ren, S., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal. Mach. Intell. 39(6), 1137–1149 (2017)
Zhang, Y., Gun, Z.Y., Wei, W.W.: Traffic sign detection based on improved faster R-CNN model. Laser Optoelectron. Prog 57(18), 173–181 (2020)
Li, X.J., Quan, L.F., Liu, D.M.: Traffic sign detection algorithm based on faster-RCNN. J. Jilin Univ. (Eng. Technol. Ed.), 2023–09–19,1–10 (2023).https://doi.org/10.13229/j.cnki.jdxbgxb.20230553
Liu, W., et al.: SSD: single shot multibox detector. In: Computer Vision--ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11--14, 2016, Proceedings, Part I 14, pp. 21–37 (2016). https://doi.org/10.1007/978-3-319-46448-0_2
Redmon, J., Divvala, S., Girshick, R.: You only look once: unified, real-time object detection. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016 December, pp. 779–788 (2016)
Wan, J.X., et al.: An efficient small traffic sign detection method based on YOLOv3. J. Sign. Process. Syst. 93(8), 899 – 911 (2020)
Cao, J., et al: An improved YOLOv4 lightweight traffic sign detection algorithm. IAENG Int. J. Comput. Sci. 50(3), 825−831(2023)
Xiao, Y.Q., Yin, S.A., Cui, G.Z., et al.: E-YOLOv4-tiny: a traffic sign detection algorithm for urban road scenarios. Frontiers neurorobotics 17, 1220443–1220443 (2023)
Li, Y.Q.: Deep learning traffic sign recognition based on improved YOLOv5, Master’s thesis, Changsha City, Hunan Province：Central South University of Forestry and Technology,2022.Bucila, C., Caruana, R., Niculescu-Mizil A., (eds.) .Model Compression[C]//ACM SIGKDD International Conference on Knowledge Discovery and Data Mining(KDD’06).Computer Science Cornell University (2006)
Hinton, G., Vinyals, O., Dean, J.: Distilling the Knowledge in a Neural Network. Comput. Sci. 14(7), 38–39 (2015)
Chen, J.R., et al.: Run, don’t walk: chasing higher FLOPS for faster neural networks. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 12021–12031(2023)
Ding. X., et al.: RepVGG: Making VGG-style convnets great again. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 13733–13742 (2021)
Ouyang, D., et al.: Efficient multi-scale attention module with cross spatial learning. In: ICASSP 2023–2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1−5. IEEE (2023)
Tan, M.X., Pang, R.M., Le, Q.V.: EfficientDet: scalable and efficient object detection. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, pp. 10778–10787 (2020)
Zheng, Z.H., et al.: Distance-IoU loss: faster and better learning for bounding box regression. Proc. AAAI Conf. Artif. Intell. 34(07), 12993–13000 (2019)
Lin, T.Y., et al.: Focal loss for dense object detection. IEEE Trans. Pattern Anal. Mach. Intell. 99, 2999–3007 (2017)
Lee, J., et al.: Layer-adaptive sparsity for the Magnitude-based Pruning (2020).https://doi.org/10.48550/arXiv.2010.07611
Shu, C.Y., et al.: Channel-wise Knowledge Distillation for Dense Prediction. In: 18th IEEE/CVF International Conference on Computer Vision (ICCV), Montreal QC, Canada, 2021, pp. 5291–5300 (2021)
Yang, L.R., et al.: Bridging cross-task protocol inconsistency for distillation in dense object detection. In: IEEE/CVF International Conference on Computer Vision (ICCV), pp. 17129–17138 (2023)
Tian, P., Mao, L.: Improved YOLOv8 object detection algorithm for the traffic sign target. Comput. Eng. Appl. 60(08), 202–212 (2024)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zhao, L., Li, D., Fang, J., Dong, X., Li, Z. (2024). YOLO-FLC: Lightweight Traffic Sign Detection Algorithm. In: Huang, DS., Zhang, C., Guo, J. (eds) Advanced Intelligent Computing Technology and Applications. ICIC 2024. Lecture Notes in Computer Science, vol 14871. Springer, Singapore. https://doi.org/10.1007/978-981-97-5609-4_7
Download citation
DOI: https://doi.org/10.1007/978-981-97-5609-4_7
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-97-5608-7
Online ISBN: 978-981-97-5609-4
eBook Packages: Computer ScienceComputer Science (R0)