Skip to main content
SpringerLink
Log in

Single image dehazing based on multi-scale segmentation and deep learning

  • Original Paper
  • Published:
Machine Vision and Applications Aims and scope Submit manuscript

Abstract

Existing image dehazing methods suffer from problems of insufficient dehazing, distortion, and low color contrast. Aiming at this problem, a deep learning single-image dehazing method based on multi-scale segmentation is proposed in this paper. The study found that the haze information in the haze image will decrease with the increase of frequency. Therefore, the haze image is first decomposed into four sub-images of different frequency domains through image segmentation in this article. A dehazing network model composed of four sub-network channels with different complexity is then constructed to extract the haze information contained in each sub-image. After the transmission sub-images are generated, the image fusion technology is used to obtain the final transmittance map. Finally, the haze-free image is obtained based on the physical model of atmospheric scattering. Experimental results on the synthetic and real images dataset show that the proposed method achieves significant dehazing effect and high color contrast with no distortion, showing superior performance than other dehazing methods.

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

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

References

  1. Cao, X.M., Liu, C.X., Zhang, J.D., et al.: Fast image dehazing algorithm based on luminance contrast enhancement. J. Comput. Aided Des. Comput. Graph. 30(10), 1925–1934 (2018)

    Google Scholar 

  2. Gao, Z.Z., Wei, W.B., Pan, Z.K., Hou, G.G., Zhao, H., Song, J.: Image dehazing based on second-order variational model. J. Comput. Aided Des. & Comput. Graph. 31(11), 1981–1994 (2019)

    Google Scholar 

  3. He, K. M., Sun, J., Tang, X. O.: Single image haze removal using dark channel prior[C]//2009. In: IEEE Conference on Computer Vision and Pattern Recognition, June 20–25, 2009, Miami, FL, USA. New York: IEEE, 2009 pp. 1956–1963

  4. He, K.M., Sun, J., Tang, X.O.: Single image haze removal using dark channel prior. IEEE Trans Pattern Anal. Mach. Intell. 33(12), 2341–2353 (2011)

    Article  Google Scholar 

  5. Lei, Q., Shi, C.J., Chen, T.T.: A single sea image dehazing method based on sky region segmentation. Comput. Eng. 41(05), 237–242 (2015)

    Google Scholar 

  6. Zeng, J.X., Yu, Y.L.: Image edge preserving and dehazing algorithm combining bilateral filtering and dark channel. J. Image Graph. 22(2), 147–153 (2017)

    Google Scholar 

  7. Kim, T.K., Paik, J.K., Kang, B.S.: Contrast enhancement system using spatially adaptive histogram equalization with temporal filtering. IEEE Trans. Consum. Electron. 44(1), 82–87 (1998)

    Article  Google Scholar 

  8. Land, E.H.: An alternative technique for the computation of the designator in the retinex theory of color vision. Nat. Acad. Sci. 83(10), 3078–3080 (1986)

    Article  Google Scholar 

  9. Mai J, Zhu Q, Wu D, et al.: Back propagation neural network dehazingIn: Proceedings of the IEEE International Conference on Robotics & Biomimetics, Bali, Indonesia: IEEE, pp. 1433–1438 (2014)

  10. Ren W, Si L, Hua Z, et al. : Single image Dehazing via multi-scale convolutional neural networks. In: Proceedings of the European Conference on Computer Vision, Springer, German, pp. 154–169 (2016)

  11. Cai, B.L., Xu, X.M., Jia, K., et al.: DehazeNet: an end-to-end system for single image haze removal. IEEE Trans. Image Process. 25(11), 5187–5198 (2016)

    Article  MathSciNet  Google Scholar 

  12. Chen D, He M, Fan Q, et al.: Gated context aggregation network for image dehazing and deraining. In: 2019 IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE, pp. 1375-1383 (2019)

  13. Zhang, X., Lou, X.P., Huang, Z.Y., et al.: A multi-scale parallel residual network for image dehazing. Opt. Tech. 46(06), 707–711 (2020)

    Google Scholar 

  14. Liu, S.Y., Zhang, Z.J., Lei, B.: Research on image dehazing algorithm based on BP-Net network structure. Opt & Optoelectron Technol. 18(06), 46–52 (2020)

    Google Scholar 

  15. Li, X.W., Yuan, T.S.: A dehazing algorithm based on cyclic generative confrontation network. J. Southwest China Normal Univ. (Nat. Sci. Ed.) 45(09), 132–138 (2020)

    Google Scholar 

  16. Cox, L.J.: Optics of the atmosphere-scattering by molecules and particles. Opt. Acta Int J. Opt. 24(7), 779 (1977)

    Article  Google Scholar 

  17. Yang, A.P., Wang, H.X., Wang, J.B., et al.: Image dehazing based on transmission fusion and multi-guided filtering. Acta Opt. Sin. 38(12), 1210001 (2018)

    Article  Google Scholar 

  18. Hu, C.S., Zhan, S., Wu, C.Z.: Image superresolution based on deep learning features. Acta Autom. Sin 43(5), 814–821 (2017)

    MATH  Google Scholar 

  19. Sun, Y. H.: Research and realization of image clearing method in haze. In: Xi'an: Xi'an University of Science and Technology, pp. 18–20 (2018)

  20. Xu, Y., Sun, M.S.: Convolution neural network image Dehazing based on multi-feature fusion. Laser Optoelectronics Progr. 55(3), 260–262 (2018)

    MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Harbin University of Science and Technology, Harbin, 150080, China

    Tianhe Yu, Ming Zhu & Haiming Chen

Authors
  1. Tianhe Yu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ming Zhu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Haiming Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Tianhe Yu.

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

Yu, T., Zhu, M. & Chen, H. Single image dehazing based on multi-scale segmentation and deep learning. Machine Vision and Applications 33, 33 (2022). https://doi.org/10.1007/s00138-022-01285-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00138-022-01285-y

Keywords

Search

Navigation