Adaptive transmission compensation via human visual system for efficient single image dehazing
- 608 Downloads
Dark channel prior has been used widely in single image haze removal because of its simple implementation and satisfactory performance. However, it often results in halo artifacts, noise amplification, over-darking, and/or over-saturation for some images containing heavy fog or large sky patches where dark channel prior is not established. To resolve this issue, this paper proposes an efficient single dehazing algorithm via adaptive transmission compensation based on human visual system (HVS). The key contributions of this paper are made as follows: firstly, two boundary constraints on transmission are deduced to preserve the intensity of the defogged image and suppress halo artifacts or noise via the minimum intensity constraint and the just-noticeable distortion model, respectively. Secondly, an improved HVS segmentation algorithm is employed to detect the saturation areas in the input image. Finally, an adaptive transmission compensation strategy is presented to remove the haze and simultaneously suppress the halo artifacts or noise in the saturation areas. Experimental results indicate that this proposed method can efficiently improve the visibility of the foggy images in the challenging condition.
KeywordsSingle image dehazing Human visual system Just-noticeable distortion Dark channel prior
This work was supported by the National High Technology Research and Development Program of China (863 Program, Grant No. 2012AA112312), National Natural Science Foundation of China (Grant No.61471166 and 61175075), the Science and Technique Project of Ministry of Transport of the People’s Republic of China (Grant No. 201231849A70) and Hunan Provincial Natural Science Foundation of China (14JJ2052).
- 6.Tan, R.T.: Visibility in bad weather from a single image. In: IEEE International Conference on Computer Vision (CVPR). New York, USA (2008)Google Scholar
- 8.He, K., Sun, J., Tang, X: Single image haze removal using dark channel prior. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1956–1963 (2009)Google Scholar
- 9.Yan, W., Bo, W.: Improved single image dehazing using dark channel prior. In: 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems (ICIS), pp. 789–792 (2010)Google Scholar
- 15.McCartney, E.J.: Optics of Atmosphere: Scattering by Molecules and Particles. Wiley, New York (1976)Google Scholar
- 16.Koschmieder, H.: Theorie der horizontaler Sichtweite Beitraege. Phys. Freib. Atmos. 12, 33–55 (1925)Google Scholar
- 19.Yan, Q., Xu, L., Jia, J.: Dense scattering layer removal. In: SIGGRAPH Asia 2013 Technical Briefs. ACM New York, NY, USAGoogle Scholar
- 20.Meng, G., Wang, Y., Duan, J., Xiang, S., Pan, C.: Efficient image dehazing with boundary constraint and contextual regularization. In: IEEE International Conference on Computer Vision (ICCV), pp. 617–624. Sydney, NSW (2013)Google Scholar
- 24.He, K., Sun, J., Tang, X.: Guided image filtering. In: The 11th European Conference on Computer Vision (ECCV), pp. 1–14. Heraklion, Crete, Greece (2010)Google Scholar
- 25.Tarel, J., Ere, N.H.: Fast visibility restoration from a single color or gray level image. In: IEEE International Conference on Computer Vision, pp. 2201–2208. New York, USA (2009)Google Scholar