Advertisement

Urban Road Water Recognition Based on Deep Learning

  • Xuefeng Tan
  • Mantao WangEmail author
  • Linchao He
  • Mengting Luo
  • Yunlu Lu
  • Jing He
  • Dejun Zhang
Conference paper
  • 7 Downloads
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 551)

Abstract

The urban road water recognition algorithm based on deep learning can achieve the automatic recognition of road water and timely report to relevant departments, thereby eliminating traffic accidents and road congestion caused by road water, and the method effectively reduces hardware resources cost. In this paper, an image recognition algorithm based on ResNet50 is proposed to accomplish the real-time intelligent recognition of urban road water. Firstly, the ResNet50 network is used to extract image features, and then the model training is carried out on the dataset of urban road water. Finally, the trained model is applied to the surveillance video of urban road to recognize the water by optimizing the model parameters. On the dataset of urban road water, the accuracy of the water state recognition of the algorithm is 94.10%, the recognition accuracy of the water level is 59.75%, and the processing speed reaches 15 FPS. The experimental results show that the algorithm has basically achieved real-time recognition of urban road water.

Keywords

Deep learning Road water Image recognition ResNet50 

Notes

Acknowledgments

This research was supported in part by the National Natural Science Foundation of China under Grant 61702350, and in part by the Youth Fund of the Sichuan Provincial Education Department under Grant 18ZB0467 and the Lab of Agricultural Information Engineering, Sichuan Key Laboratory.

References

  1. 1.
    Pengfei, S., Yanwei, Z., Mingxia, Y.: Research on urban waterlogging monitoring and pre-warning system. Comput. Meas. Control 24(02), 49–52 (2016). (in Chinese)Google Scholar
  2. 2.
    Xia, Q., Bowang, G., Yanchao, C.: Urban road water monitoring system based on cloud server. Microcontrollers & Embedded Systems 15(11), 37–39+43 (2015). (in Chinese)Google Scholar
  3. 3.
    Hong-tu, B.A.I., Hua, L.I.U.: Design of road seeper monitor based on STM32 singlechip. Tech. Autom. Appl. 37(04), 156–158 (2018). (in Chinese)Google Scholar
  4. 4.
    Qinglin, X., Haipeng, W., Fei, Z.: Research on monitoring equipment of road water. Shanxi Sci. Technol. 31(06), 74–77 (2016). (in Chinese)Google Scholar
  5. 5.
    Krizhevsky, A., Sutskever, I., Hinton, G.E., et al.: ImageNet classification with deep convolutional neural networks. In: Neural Information Processing Systems, pp. 1097–1105 (2012)Google Scholar
  6. 6.
    Toshev, A., Szegedy, C.: DeepPose: human pose estimation via deep neural networks. In: Computer Vision and Pattern Recognition, pp. 1653–1660 (2014)Google Scholar
  7. 7.
    Long, J., Shelhamer, E., Darrell, T., et al.: Fully convolutional networks for semantic segmentation. In: Computer Vision and Pattern Recognition, pp. 3431–3440 (2015)Google Scholar
  8. 8.
    Ren, S., He, K., Girshick, R.B., 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)CrossRefGoogle Scholar
  9. 9.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: International Conference on Learning Representations (2015)Google Scholar
  10. 10.
    Szegedy, C., Liu, W., Jia, Y., et al.: Going deeper with convolutions. In: Computer Vision and Pattern Recognition, pp. 1–9 (2015)Google Scholar
  11. 11.
    He, K., Zhang, X., Ren, S., et al.: Deep residual learning for image recognition. In: Computer Vision and Pattern Recognition, pp. 770–778 (2016)Google Scholar
  12. 12.
    Szegedy, C., Ioffe, S., Vanhoucke, V., et al.: Inception-v4, Inception-ResNet and the impact of residual connections on learning. In: National Conference on Artificial Intelligence, pp. 4278–4284 (2016)Google Scholar
  13. 13.
    Howard, A.G., Zhu, M., Chen, B., et al.: MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv: Computer Vision and Pattern Recognition (2017)Google Scholar
  14. 14.
    Hu, J., Shen, L., Sun, G., et al.: Squeeze-and-Excitation Networks. In: Computer Vision and Pattern Recognition, pp. 7132–7141 (2018)Google Scholar
  15. 15.
    Zhang, X., Zhou, X., Lin, M., et al.: ShuffleNet: an extremely efficient convolutional neural network for mobile devices. In: Computer Vision and Pattern Recognition, pp. 6848–6856 (2018)Google Scholar
  16. 16.
    Suhao, L., Jinzhao, L., Guoquan, L., et al.: Vehicle type detection based on deep learning in traffic scene. Procedia Comput. Sci. 131, 564–572 (2018)CrossRefGoogle Scholar
  17. 17.
    Yong, F., Qian, D., Jian, W., Weifeng, W., Aiwei, C.: Road traffic congestion detection based on deep learning. Intell. City 4(23), 1–3 (2018). (in Chinese)Google Scholar
  18. 18.
    Qunfang, X., Jun, L., Wei, Y.: A method of fatigue driving state detection based on deep learning. Control Inf. Technol. 06, 91–95 (2018). (in Chinese)Google Scholar
  19. 19.
    Shi-Qi, D., Zhi-Yong, Z.: Fatigue driving detection algorithm based on deep learning. Comput. Syst. Appl. 27(07), 113–120 (2018). (in Chinese)Google Scholar
  20. 20.
    Haqiang, X., Zhiqi, D., Bo, S.: Pedestrian detection method based on modified SSD. Comput. Eng. 44(11), 228–233+238 (2018). (in Chinese)Google Scholar
  21. 21.
    Lin, M., Chen, Q., Yan, S., et al.: Network in network. Int. Conf. Learn. Represent. (2014)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  • Xuefeng Tan
    • 1
  • Mantao Wang
    • 1
    • 2
    Email author
  • Linchao He
    • 1
  • Mengting Luo
    • 1
  • Yunlu Lu
    • 1
  • Jing He
    • 3
  • Dejun Zhang
    • 4
  1. 1.College of Information and EngineeringSichuan Agricultural UniversityYaanChina
  2. 2.The Lab of Agricultural Information Engineering, Sichuan Key LaboratoryYaanChina
  3. 3.College of Information Science and TechnologyChengdu University of TechnologyChengduChina
  4. 4.School of Geography and Information EngineeringChina University of GeosciencesWuhanChina

Personalised recommendations