Single Image Dehazing Algorithm Based on Sky Region Segmentation

  • Weixiang LiEmail author
  • Wei JieEmail author
  • Somaiyeh MahmoudzadehEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11888)


In this paper a hybrid image defogging approach based on region segmentation is proposed to address the dark channel priori algorithm’s shortcomings in de-fogging the sky regions. The preliminary stage of the proposed approach focuses on segmentation of sky and non-sky regions in a foggy image taking the advantageous of Meanshift and edge detection with embedded confidence. In the second stage, an improved dark channel priori algorithm is employed to defog the non-sky region. Ultimately, the sky area is processed by DehazeNet algorithm, which relies on deep learning Convolutional Neural Networks. The simulation results show that the proposed hybrid approach in this research addresses the problem of color distortion associated with sky regions in foggy images. The approach greatly improves the image quality indices including entropy information, visibility ratio of the edges, average gradient, and the saturation percentage with a very fast computation time, which is a good indication of the excellent performance of this model.


Image haze removal Atmospheric scattering model Regional segmentation Dark channel prior DehazeNet 


  1. 1.
    Dan-Dan, C., Li, C., Yong-Xin, Z.: Fog removal of single image with modified atmospheric dissipation function. Chin. J. Image Graph. 22(06), 787–796 (2017)Google Scholar
  2. 2.
    Ping, W., Yunfeng, Z., Fangxun, B., Hongwei, D., Caiming, Z.: Optimization of fog removal method based on image quality reduction model in foggy days. Chin. J. Image Graph. 23(04), 605–616 (2008)Google Scholar
  3. 3.
    Yufei, D., Yan, Y., Biting, C.: Single image depth defogging algorithm based on guidance graph optimization. Comput. Appl. 37(01), 268–277 (2017)Google Scholar
  4. 4.
    Yang, L., Jie, Z., Hui, Z.: A modified Retinex algorithm for image defogging research and application. Comput. Sci. 45(S1), 242–251 (2018)Google Scholar
  5. 5.
    Wenchao, J., Haibo, L., Yujie, Y., Jiafeng, C., Aobing, S.: A high similarity image recognition and classification algorithm combining wavelet transform and convolutional neural network. Comput. Eng. Sci. 40(09), 1646–1652 (2018)Google Scholar
  6. 6.
    Dian-Wei, W., Wei-Chao, Y., Ying, L., Ting-Ge, Z.: Fast single image defogging algorithm based on dark channel prior. Comput. Appl. Res. 01, 1–7 (2019)Google Scholar
  7. 7.
    Tan, R.T.: Visibility in bad weather from a single image. In: IEEE Conference on Computer Vision and Pattern Recognition, Washington, pp. 1–8 (2008)Google Scholar
  8. 8.
    Fattal, R.: Single Image Dehazing. ACM Trans. Graph. 27(3), 72 (2008)CrossRefGoogle Scholar
  9. 9.
    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)CrossRefGoogle Scholar
  10. 10.
    He, K.M., Sun, J., Tang, X.O.: Guided image filtering. IEEE Trans. Pattern Anal. Mach. Intell. 35(6), 1397–1409 (2013)CrossRefGoogle Scholar
  11. 11.
    Jieping, L., Bingkun, H., Gang, W.: A fast algorithm for defogging single image. Acta Electronica Sinica 45(08), 1896–1901 (2017)Google Scholar
  12. 12.
    Li, B., Peng, X., Wang, Z.: An All-in-One Network for Dehazing and Beyond. arXiv: Computer Vision and Pattern Recognition (2017)Google Scholar
  13. 13.
    Zhang, D., Ju, M., Wang, X.: A fast image defogging algorithm based on dark channel prior. Acta Electronica Sinica 43(07), 1437–1443 (2015)Google Scholar
  14. 14.
    Cai, B., Xu, X., Jia, K., Qing, C., Tao, D.: DehazeNet: an end-to-end system for single image haze removal. IEEE Trans. Image Process. 25(11), 5187–5198 (2016)MathSciNetCrossRefGoogle Scholar
  15. 15.
    Chengming, R.: Research on defogging algorithm for single fog image. Anhui University (2017)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Nanjing Tech UniversityNanjingChina
  2. 2.Deakin UniversityGeelongAustralia

Personalised recommendations