Learning deep transmission network for efficient image dehazing

Abstract

Single image dehazing algorithms are recently attracting more and more attention from many researchers because of their flexibility and practicality. However, most existing algorithms have some challenges in dealing with images captured under complex weather conditions because the often used assumptions cannot always reflect true structural information of natural images in those situations. In this paper, we develop a deep transmission network to estimate the transmission map for efficient image dehazing, which automatically explores and exploits underlying haze-relevant features from RGB color channels and a local patch jointly for robust transmission estimation. Moreover, due to the fact that transmission values are affected by light wavelengths, a three-channel transmission map is considered in the proposed network so that this network can discover and utilize the chromatic characteristics for transmission estimation. We also investigate different network structures and parameter settings to achieve different trade-offs between performance and speed, and find that three color channels and local spatial information are the most informative haze-relevant features. This could explain why haze relevant priors or assumptions are often related to three color channels in most existing methods. Experiment results demonstrate that the proposed algorithm outperforms state-of-the-art methods on both synthetic and real-world datasets.

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

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13

References

  1. 1.

    Aharon M, Elad M, Bruckstein A (2006) The K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation. IEEE Trans Signal Proccess 54 (11):4311–4322

    Article  Google Scholar 

  2. 2.

    Ancuti CO, Ancuti C (2013) Single image dehazing by multi-scale fusion. IEEE Trans Image Process 22(8):3271–3282

    Article  Google Scholar 

  3. 3.

    Ancuti C, Ancuti CO, De Vleeschouwer C (2016) D-hazy: a dataset to evaluate quantitatively dehazing algorithms. In: IEEE International conference on image processing (ICIP), pp 2226–2230

  4. 4.

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

    MathSciNet  Article  Google Scholar 

  5. 5.

    Choi LK, You J, Bovik AC (2015) Referenceless prediction of perceptual fog density and perceptual image defogging. IEEE Trans Image Process 24(11):3888–3901

    MathSciNet  Article  Google Scholar 

  6. 6.

    Danescu R, Pantilie C et al (2012) Particle grid tracking system stereovision based obstacle perception in driving environments. IEEE Intell Transp Syst Mag 4 (1):6–20

    Article  Google Scholar 

  7. 7.

    Dong C, Loy CC, He K, Tang X (2016) Image super-resolution using deep convolutional networks. IEEE Trans Pattern Anal Mach Intell 38(2):295–307

    Article  Google Scholar 

  8. 8.

    Fattal R (2008) Single image dehazing. ACM Trans Graph 27(3):721–729

    Article  Google Scholar 

  9. 9.

    Fattal R (2014) Dehazing using color-lines. ACM Trans Graph 34(1):1–13

    Article  Google Scholar 

  10. 10.

    Gaikwad V, Lokhande S (2015) Lane departure identification for advanced driver assistance. IEEE Trans Intell Transp Syst 16(2):910–918

    Google Scholar 

  11. 11.

    Grabner M, Kvicera V (2011) The wavelength dependent model of extinction in fog and haze for free space optical communication. Opt Express 19(4):3379–3386

    Article  Google Scholar 

  12. 12.

    Hao J, Li C, Kim Z, Xiong Z (2013) Spatio-temporal traffic scene modeling for object motion detection. IEEE Trans Intell Transp Syst 14(1):295–302

    Article  Google Scholar 

  13. 13.

    Hautière N, Tarel J, Aubert D, Dumont (2008) Blind contrast enhancement assessment by gradient ratioing at visible edges. Image Analy Stereology 27(2):87–95

    MathSciNet  Article  Google Scholar 

  14. 14.

    Hautiere N, Tarel JP et al (2010) Mitigation of visibility loss for advanced camera-based driver assistance. IEEE Trans Intell Transp Syst 11(2):474–484

    Article  Google Scholar 

  15. 15.

    He K, Sun J (2015) Convolutional neural networks at constrained time cost. In: Internal conference on computer vision and pattern recognition (CVPR), pp 5353–5360

  16. 16.

    He K, Sun J, Tang X (2009) Single image haze removal using dark channel prior. In: IEEE Conference on computer vision and pattern recognition (CVPR), pp 1956–1963

  17. 17.

    Huang S, Chen B et al (2014) Visibility restoration of single hazy images captured in real-world weather conditions. IEEE Trans Circuits Syst Video Tech 24 (10):1814–1824

    Article  Google Scholar 

  18. 18.

    Kim D, Jeon C, Kang B, Ko H (2008) Enhancement of image degraded by fog using cost function based on human visual model. In: IEEE International conference on multisensor fusion and integration for intelligent systems. Seoul

  19. 19.

    Kopf J, Neubert B, Chen B, Cohen MF, Deussen O et al (2008) Konstanz deep photo: model-based photograph enhancement and viewing. ACM Trans Graphics 27(5):116,1–116,10

    Article  Google Scholar 

  20. 20.

    Koschmieder H (1925) Theorie der horizontaler Sichtweite Beitraege. Physicae Freiberger Atmosphere, 33–55

  21. 21.

    Kruse PW, McGlauchlin LD, McQuistan RB (1962) Elements of infrared technology: generation, transmission and detection. Wiley, New York. Chap. 5

    Google Scholar 

  22. 22.

    Lai Y, Chen Y, Chiou C, Hsu C (2015) Single-image dehazing via optimal transmission map under scene priors. IEEE Trans Circuits Syst Video Tech 25(1):1–14

    Article  Google Scholar 

  23. 23.

    LeCun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278–2324

    Article  Google Scholar 

  24. 24.

    Lu H, Li Y, Nakashima S, Serikawa S (2016) Single image dehazing through improved atmospheric light estimation. Multimed Tools Appl 75(24):17081–17096

    Article  Google Scholar 

  25. 25.

    Meng G, Wang Y, Duan J, Xiang S, Pan C (2013) Efficient image dehazing with boundary constraint and contextual regularization. In: IEEE International conference on computer vision (ICCV), pp 617–624

  26. 26.

    Narasimhan SG, Nayar SK (2002) Vision and the atmosphere. Int J Comput Vis 48(3):233–254

    Article  Google Scholar 

  27. 27.

    Narasimhan SG, Nayar SK (2003) Contrast restoration of weather degraded images. IEEE Trans Pattern Anal Mach Intell 25(6):713–724

    Article  Google Scholar 

  28. 28.

    Narasimhan SG, Nayar SK (2003) Interactive (de)weathering of an image using physical models. IEEE workshop on color and photometric methods in computer vision. In: Conjunction with ICCV

  29. 29.

    Nayar SK, Narasimhan SG (1999) Vision in bad weather. In: The Seventh IEEE international conference on computer vision. Kerkyra

  30. 30.

    Oakley JP, Bu H (2007) Correction of simple contrast loss in color images. IEEE Trans Image Process 16(2):511–522

    MathSciNet  Article  Google Scholar 

  31. 31.

    Ren W, Si L, Hua Z, Jinshan P, Xiaochun C, Yang M (2016) Single image dehazing via multi-scale convolutional neural networks. In: The 14th European conference on computer vision (ECCV), pp 154–169

  32. 32.

    Rong Z, Jun WL (2014) Improved wavelet transform algorithm for single image dehazing. Optik - Int J Light Electron Optic 125(13):3064–3066

    Article  Google Scholar 

  33. 33.

    Schechner YY, Narasimhan SG, Nayar SK (2003) Polarization-based vision through haze. Appl Opt 42(3):511–525

    Article  Google Scholar 

  34. 34.

    Tan RT (2008) Visibility in bad weather from a single image. In: IEEE International conference on computer vision (CVPR), pp 2347–2354

  35. 35.

    Tan H, He X, Wang Z, Liu G (2017) Parallel implementation and optimization of high definition video real-time dehazing. Multimed tools Appl 76(22):23413–23434

    Article  Google Scholar 

  36. 36.

    Tang K, Yang J, Wang J (2014) Investigating haze-relevant features in a learning framework for image dehazing. In: The International conference computer vision and pattern recognition (CVPR), pp 2995–3002

  37. 37.

    Tarel JP, Hautiere N, Caraffa L, Cord A, Halmaoui H, Gruyer D (2012) Vision enhancement in homogeneous and heterogeneous fog. IEEE Intell Transp Syst Mag 4(2):6–20

    Article  Google Scholar 

  38. 38.

    Tian B, Li Y, Li B, Wen D (2014) Rear-view vehicle detection and tracking by combining multiple parts for complex urban surveillance. IEEE Trans Intell Transp Syst 15(2):597–606

    Article  Google Scholar 

  39. 39.

    Wu BF, Kao C, Juang J, Huang Y (2013) A new approach to video-based traffic surveillance using fuzzy hybrid information inference mechanism. IEEE Trans Intell Transp Syst 14(1):485–491

    Article  Google Scholar 

  40. 40.

    Xu Y, Xu D et al (2012) Detection of sudden pedestrian crossings for driving assistance systems. IEEE Trans Syst Man Cybern B Cybern 42(3):729–739

    Article  Google Scholar 

  41. 41.

    Xu Z, Liu X, Chen X (2009) Fog removal from video sequences using contrast limited adaptive histogram equalization. In: International conference on computational intelligence and software engineering, pp 1–4

  42. 42.

    Zhou J, Zhou F (2013) Single image dehazing motivated by Retinex theory. In: The 2nd International symposium on instrumentation and measurement, sensor network and automation (IMSNA)

  43. 43.

    Zhou W, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13 (4):600–612

    Article  Google Scholar 

  44. 44.

    Zhu Q, Mai J, Shao L (2015) A fast single image haze removal algorithm using color attenuation prior. IEEE Trans Image Process 24(11):3522–3533

    MathSciNet  Article  Google Scholar 

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No. 61471166, 61471167 and 61671204) and Natural Science Foundation of Hunan Province (CN) (14JJ2052).

Author information

Affiliations

Authors

Corresponding author

Correspondence to Zhigang Ling.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ling, Z., Fan, G., Gong, J. et al. Learning deep transmission network for efficient image dehazing. Multimed Tools Appl 78, 213–236 (2019). https://doi.org/10.1007/s11042-018-5687-0

Download citation

Keywords

  • Image dehazing
  • Haze-relevant features
  • Convolutional neural networks
  • Deep transmission network