Skip to main content
SpringerLink
Log in

A Combined Multi-Mode Visibility Detection Algorithm Based on Convolutional Neural Network

  • Published:
Journal of Signal Processing Systems Aims and scope Submit manuscript

Abstract

The accuracy of visibility detection greatly affects daily life and traffic safety. Existing visibility detection methods based on deep learning rely on massive haze images to train neural networks to obtain detection models, which are prone to overfit in dealing with small samples cases. In order to overcome this limitation, a large amount of measured data are used to train and optimize the convolutional neural network, and an improved DiracNet method is proposed to improve the accuracy of the algorithm. On this foundation, combined multi-mode algorithm is proposed to achieve small samples fitting and train an effective model in a short time. In this paper, the proposed improved DiracNet and the combined multi-mode algorithm are verified by using the measured atmospheric fine particle concentration data (pm1.0, pm2.5, pm10) and haze video data. The validation results demonstrate the effectiveness of the proposed algorithm.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6

Similar content being viewed by others

References

  1. Graves, N., & Newsam, S. (2011). Using visibility cameras to estimate atmospheric light extinction, Applications of Computer Vision (WACV). Kona: IEEE.

  2. Varjo, S., Kaikkonen, V., Hannuksela, J., et al. (2015). All-in-focus image reconstruction from in-line holograms of snowflakes, 2015 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) Proceedings, 2015. Pisa: IEEE.

  3. Li, G., Wu, J. F., & Lei, Z. Y. (2014). Research progress on image smog grade evaluation and defogging technology. Laser Journal, 9(1), 1–6.

    Google Scholar 

  4. Cheng, X., Yang, B., Liu, G., et al. (2018a). A total bounded variation approach to Low visibility estimation on expressways. Sensors, 18(2), 392–392.

    Google Scholar 

  5. Zhou, J. (2017). Study of visibility detection algorithm based on traffic image. Techniques of Automation and Applications, 36(10), 100–103.

    Google Scholar 

  6. Chen, Z., Yan, Z., & Yao, Z. (2019). Visibility measurement using dark channel prior method based on adaptive fog concentration coefficient. Modern Electronics Technique, 42(9), 39–45.

    Google Scholar 

  7. Liu, J., & Liu, X. (2015). Visibility estimation algorithm for fog weather based on inflection point line. Journal of Computer Applications, 35(2), 528–530.

    Google Scholar 

  8. Chen, Z. (2016). PTZ visibility detection based on image luminance changing tendency. International Conference on Optoelectronics & Image Processing. IEEE.

  9. Hautiere, N., Labayrade, R., & Aubert, D. (2006a). Real-Time Disparity Contrast Combination for Onboard Estimation of the Visibility Distance. IEEE Transactions on Intelligent Transportation Systems, 7(2), 201–212.

    Article  Google Scholar 

  10. Babari, R., Hautiere, N., Dumont, £., et al. (2011). A model-driven approach to estimate atmospheric visibility with ordinary cameras. Atmospheric Environment, 45(30), 5316–5324.

  11. Cheng, X., Yang, B., Liu, G., et al. (2018b). A variational approach to atmospheric visibility estimation in the weather of fog and haze. Sustainable Cities & Society, 39(1), 215–224.

    Article  Google Scholar 

  12. Wang, T. (2019). Research on haze visibility detection based on spectrum analysis and improved inception. Nanjing University of Posts and Telecommunications.

  13. Hongjun, L. V. (2019). Research on fog visibility detection algorithms based on deep learning. Nanjing University of Posts and Telecommunications.

  14. Hautiére, N., Tarel, J. P., Lavenant, J., et al. (2006b). Automatic fog detection and estimation of visibility distance through use of an onboard camera. Machine Vision and Applications, 17(1), 8–20.

    Article  Google Scholar 

  15. Koschmieder, H. (1924). Theorie der horizontalen Sichtweite: Kontrast und Sichtweite. Keim & Nemnich, pp. 171–181.

  16. He, K., Zhang, X., Ren, S., et al. (2016). Deep Residual Learning for Image Recognition, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE Computer Society.

  17. Zagoruyko, S., & Komodakis, N. (2018) DiracNets: Training Very Deep Neural Networks Without Skip-Connections, 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. Salt Lake: IEEE Computer Society.

  18. Zamir, A., Sax, A., Shen, W., et al. (2018). Taskonomy: Disentangling Task Transfer Learning, 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. Salt Lake: IEEE Computer Society.

Download references

Acknowledgements

This work is supported by the opening fund for State Key Laboratory of Severe Weather of China Meteorological Administration (Grant No.2021LASW-A07), the Universities Natural Science Research project of Jiangsu Province (Grant No.19KJB510048), the Opening fund for National Key Laboratory of Solid Microstructure Physics in Nanjing University (Grant No.M30006), the Post-doctoral fund of Jiangsu Province (Grant No.1701132B).

Author information

Authors and Affiliations

  1. Key Laboratory of Transportation Meteorology, China Meteorological Administration, Nanjing, Jiangsu, China

    Mu Xiyu, Xu Qi, Zhang Qiang, Wang Hongbin & Zhou Linyi

  2. Nanjing Joint Institute for Atmospheric Sciences, Nanjing, Jiangsu, China

    Mu Xiyu, Wang Hongbin & Zhou Linyi

  3. Jiangsu Air Traffic Management Branch Bureau of CAAC, Nanjing, Jiangsu, China

    Xu Qi

  4. Meteorology Center of North China Air Traffic Management Bureau, CAAC, Beijing, China

    Zhang Qiang

  5. Jiangsu Telecommunication Limited Company, Pukou Branch, Jiangsu, Nanjing, China

    Ren Junch

  6. Nanjing University of Posts and Telecommunications, Nanjing, Jiangsu, China

    Ren Junch

Authors
  1. Mu Xiyu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xu Qi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Zhang Qiang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Ren Junch
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Wang Hongbin
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Zhou Linyi
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Mu Xiyu.

Ethics declarations

Conflict of Interest

The authors have no competing interests to declare that are relevant to the content of this article.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Xiyu, M., Qi, X., Qiang, Z. et al. A Combined Multi-Mode Visibility Detection Algorithm Based on Convolutional Neural Network. J Sign Process Syst 95, 49–56 (2023). https://doi.org/10.1007/s11265-022-01792-1

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11265-022-01792-1

Keywords

Search

Navigation