Haze Removal Algorithm Using Improved Restoration Model Based on Dark Channel Prior

  • Dai Zhen
  • Hamid A. JalabEmail author
  • Liu Shirui
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11870)


In recent years, with the rapid development of social economy and the people’s living standard, the awareness of security precaution is becoming increasingly important. However, in severe weather conditions, rain and haze have a large influence on the images obtained by video monitoring like the image contrast information, and it has a bad impact on the security work. So the clarity of images becomes very meaningful, and researchers start to pay attention to the field of image dehazing. Among many studies, the dark channel prior dehazing algorithm is a major breakthrough in the field of image dehazing technology. The advantages of this algorithm mainly focus on solving the problem of over-dependence on the physical model of sky scattering, poor adaptability to images containing sky regions and the problem that the processed images are too dark. Based on the atmospheric physical model, this research proposed a haze removal algorithm based on dark channel prior, which has good robustness to the bright sky regions, and also has a good effect on the edges. Firstly, the haze images were detected to see whether sky regions are included or not. Secondly, haze images that have sky regions were segmented into sky region and non-sky region. Then, the haze images that have sky regions and that have no sky regions were dehazed respectively. Finally, the quantitative assessment is based on NIQE (Natural Image Quality Evaluator), and the qualitative assessment is based on human subjective visual standards to compare the intrinsic images results. Quantitative and qualitative results both demonstrate that the performance of the proposed method is better than other techniques in the field of image dehazing that using dark channel prior. The proposed method can be better applied in the area of video image processing.


Image dehazing Dark channel prior Restoration model Image quality assessment 


  1. 1.
    Dong, Z.H.: Image haze removal improved algorithm using dark channel prior. M.S. thesis. Lanzhou Jiaotong University, China (2017)Google Scholar
  2. 2.
    Guo, P., Cai, Z.X., et al.: Overview and prospect of image dehazing technology. Comput. Appl. 09(3), 675–689 (2010)Google Scholar
  3. 3.
    Sheng, H.L., Isa, N.A.M., Chen, H.O., et al.: A New histogram equalization method for digital image enhancement and brightness preservation. Signal Image Video Process. 09(3), 675–689 (2015)CrossRefGoogle Scholar
  4. 4.
    Hitam, M.S., Yussof, W.N.J.H.W., Awalludin, E.A., et al.: Mixture contrast limited adaptive histogram equation for underwater image enhancement. In: International Conference on Computer Applications Technology, pp. 1–5. IEEE (2013)Google Scholar
  5. 5.
    Kim, J.Y., Kim, L.S., Hwang, S.H.: An advanced contrast enhancement using partially overlapped sub-block histogram equation. IEEE Trans. Circuits Syst. Video Technol. 11(4), 475–484 (2001)CrossRefGoogle Scholar
  6. 6.
    Russo, F.: An image enhancement technique combining sharping and noise reduction. IEEE Trans. Instrum. Measur. 51(4), 824–828 (2002)CrossRefGoogle Scholar
  7. 7.
    Tang, L., Chen, S., Liu, W., et al.: Improved retinex image enhancement algorithm. Procedia Environ. Sci. 11(Part A), 208–212 (2011)CrossRefGoogle Scholar
  8. 8.
    Fu, X., Sun, Y., Liwang, M., et al.: A novel retinex based approach for image enhancement with illumination adjustment, pp. 1190–1194 (2014)Google Scholar
  9. 9.
    Zhang, G., Yan, P., Zhao, H., et al.: A survey of image enhancement algorithms based on retinex theory. In The Third International Conference on Computer Science & Education (2008)Google Scholar
  10. 10.
    Wang, Y.C.: Research on haze removal using dark channel prior. M.S. thesis, Dalian University of Technology, China (2011)Google Scholar
  11. 11.
    Tan, R.T.: Visibility in bad weather from a single image. In: Computer Vision and Pattern Recognition (CVPR), pp. 1–8 (2008)Google Scholar
  12. 12.
    Fattal, R.: Single image dehazing. ACM Trans. Graph. 27(3), 721–729 (2008)CrossRefGoogle Scholar
  13. 13.
    He, K.M., Sun, J., Tang, X.O.: Single image haze removal using dark channel prior. IEEE Trans. Pattern Anal. Mach. Intell. 33, 2341–2353 (2011)CrossRefGoogle Scholar
  14. 14.
    Narasimhan, S.G., Nayar, S.K.: Chromatic framework for vision in bad weather. In: Proceeding of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Hilton Head Island, SC, USA, pp. 1598–1605 (2000)Google Scholar
  15. 15.
    Otsu, N.: A threshold selection method from gray-level histogram. IEEE Trans. Syst. Man Cybern. 9(1), 62–66 (2007)MathSciNetCrossRefGoogle Scholar
  16. 16.
    Wu, Y.H., Pan, C., Wu, N.: Research on improved otsu recursively segmented single image dehazing algorithm. J. XI’an Univ. Sci. Technol. 1, 438–444 (2017)Google Scholar
  17. 17.
    Mao, T.Y.: Optimizing technology of dark channel prior dehazing based on sky region segmentation. Nanjing University of Aeronautics and Astronautics (2017)Google Scholar
  18. 18.
    Mittal, A., Soundararajan, R., Bovik, A.C.: Making a completely blind image quality analyzer. IEEE Signal Process. Lett. 22(3), 209–212 (2013)CrossRefGoogle Scholar
  19. 19.
    Shanthi, S.A., Sulochana, C.H., Jerome, S.A.: Image denoising using bilateral in subsampled pyramid and nonsubsampled directional filter bank domain. J. Intell. Fuzzy Syst. 31(1), 237–247 (2016)CrossRefGoogle Scholar
  20. 20.
    Papari, G., Idowu, N., Varslot, T.: Fast bilateral filtering for denoising large 3D images. IEEE Trans. Image Process. 26(1), 251–261) (2017)MathSciNetCrossRefGoogle Scholar
  21. 21.
    Xiao, S.B., Li, Y.: Haze removal algorithm based on fast multi-scale retinex with color fidelity. Comput. Eng. Appl. 51(06), 176–180 (2015)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.University of MalayaKuala LumpurMalaysia

Personalised recommendations