Abstract
The target detection algorithm based on the YOLO-v4 model can realize real-time traffic signs because of high detection accuracy and fast detection speed. Owing to the lack of sufficient computing resources and storage space in vehicles and other embedded devices, it is difficult to deploy this kind of computing model with high requirements for power consumption. With the continuous development of intelligent transportation technology and the rise of advanced driving assistance systems. This paper proposes an improved lightweight traffic sign detection model called M-Yolo (micro-Yolo) to solve this problem. First, the feature extraction network of this model is changed from CSPDarknet53 to MobileNetv3-large network structure. Then, the SiLU activation function is used instead of the ReLU activation function in the MobileNetv3 shallow network, which further improves the detection accuracy of the algorithm and optimizes the convergence effect of model training. Finally, use the Gaussian loss function as the positioning loss of the bounding box. Experimental results show that the algorithm can guarantee real-time performance and detection accuracy and perform real-time detection tasks for embedded devices with low computing power and low storage capacity.
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
References
Zhu, Z., Liang, D., Zhang, S., et al.: Traffic sign detection and classification i-n the wild. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition, pp. 2110–2118 (2016)
Li, J., Wang, Z.: Real-time traffic sign recognition based on efficient CNNs in the wild. IEEE Trans. Intell. Transp. Syst. 20(3), 975–984 (2019)
Rezatofighi, H., Tsoi, N., Gwak, J., et al.: Generalized inter-section over union: a metric and a loss for bounding box regression. arXiv:1902.09630 (2019)
Wang, X., Xiao, T., Jiang, Y., et al.: Repulsion loss: detecting pedestrians in a crowd. In: 2018 IEEEConference on Computer Vision and Pattern Recognition, pp. 7774–7783 (2018)
Jiang, B., Luo, R., Mao, J., et al.: Acquisition of localization confidence for accurate object detection. arXiv:1807.11590 (2018)
Cai, Z., Vasconcelos, N.: Cascade R-CNN: delving into high quality object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 6154–6162 (2018)
Tabernik, D., Skocaj, D.: Deep learning for large-scale traffic-sign detection and recognition. IEEE Trans. Intell. Transp. Syst. (99), 1–14 (2019)
Lin, T., Dollár, P., Girshick, R., et al.: Feature pyramid net-works for object detection. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, pp. 936–939 (2017)
Ren, S., He, K., Girshick, R., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal.-d Mach. Intell. 39(6), 1137–1149 (2017)
Iandola, F.N., Han, S., Moskewicz, M.W., et al.: SqueezeNet: AlexNet-level accuracy with50 x fewer parameters and <0.5 MB model size. arXiv:1602.07360 (2016)
Howard, A.G., Zhu, M., Chen, B., et al.: Mobilenets: efficient convolutional neural networks for mobile vision applications. arXiv:1704.04861 (2017)
Jacob, B., Kligys, S., Chen, B., et al.: Quantization and training of neural networks for efficient integer-arithmetic-only inference (2017)
Jouppi, N.P., Young, C., Patil, N., et al.: In-datacenter performance analysis of a tensor processing unit. In: 2017 ACM/IEEE 44th Annual International Symposium (2017)
Rastegari, M., Ordonez, V., Redmon, J., et al.: XNOR-net: ImageNet classification using binary convolutional neural networks. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) Computer Vision – ECCV 2016. ECCV 2016. Lecture Notes in Computer Science, vol. 9908, pp. 525–554. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46493-0_32
Redmon, J., Farhadi, A.: Yolov3: an incremental improvement. arXiv:1804.02767 (2018)
Wu, X., Tian, Q.: Survey of traffic sign recognition. Comput. Eng. Appl. 056(10), 20–26 (2020)
Bochkovskiy, A., Wang, C.Y., Liao, H.Y.M.: YOLOv4: optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934 (2020)
Lu, J., He, J., Li, Z., et al.: A Surve of target detection based on deep learning. Electron. Opt. Control 27(5), 60–67 (2020)
Tan, M., Pang, R., Le, Q.V.: EfficientDet: scalable and efficient object detection. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 10781–10790 (2020)
Zhang, X., Zhou, X., Lin, M., et al.: Shufflenet: an extremely efficient convolutional neural network for mobile devices. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 6848–6856 (2018)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Wenze, Y. et al. (2022). An Improved Lightweight Traffic Sign Detection Model for Embedded Devices. In: Hassanien, A.E., Snášel, V., Chang, KC., Darwish, A., Gaber, T. (eds) Proceedings of the International Conference on Advanced Intelligent Systems and Informatics 2021. AISI 2021. Lecture Notes on Data Engineering and Communications Technologies, vol 100. Springer, Cham. https://doi.org/10.1007/978-3-030-89701-7_14
Download citation
DOI: https://doi.org/10.1007/978-3-030-89701-7_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-89700-0
Online ISBN: 978-3-030-89701-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)