DOI QR코드

DOI QR Code

Nonlinear model for estimating depth map of haze removal

안개제거의 깊이 맵 추정을 위한 비선형 모델

  • Lee, Seungmin (Dept. of Electronics Engineering, Dong-A University) ;
  • Ngo, Dat (Dept. of Electronics Engineering, Dong-A University) ;
  • Kang, Bongsoon (Dept. of Electronics Engineering, Dong-A University)
  • Received : 2020.05.29
  • Accepted : 2020.06.19
  • Published : 2020.06.30

Abstract

The visibility deteriorates in hazy weather and it is difficult to accurately recognize information captured by the camera. Research is being actively conducted to remove haze so that camera-based applications such as object localization/detection and lane recognition can operate normally even in hazy weather. In this paper, we propose a nonlinear model for depth map estimation through an extensive analysis that the difference between brightness and saturation in hazy image increases non-linearly with the depth of the image. The quantitative evaluation(MSE, SSIM, TMQI) shows that the proposed haze removal method based on the nonlinear model is superior to other state-of-the-art methods.

안개가 낀 악조건의 날씨에서는 가시성이 저하되어 카메라로 포착한 정보들을 정확히 인식하기 어렵다. 안개 낀 날씨에서도 사물인식, 차선 인식 등 카메라 기반의 기기들이 정상 동작할 수 있도록 안개제거 연구가 활발히 진행되고 있다. 본 논문에서는 안개 영상에서 밝기와 채도의 차이가 영상의 깊이에 따라 비선형적으로 증가한다는 분석을 통해 깊이 맵 추정을 위한 비선형 모델을 제시한다. 비선형 모델의 안개 제거 방법은 여러 가지 안개제거 방법과의 정량적 수치평가(MSE, SSIM, TMQI)를 통해 동등 이상의 결과를 보여줌으로써 우수한 성능을 자랑한다.

Keywords

References

  1. P. Xia and X. Liu, "Image dehazing technique based on polarimetric spectral analysis," Optik, vol.127, no.18, pp.7350-7358, 2016. DOI: 10.1016/j.ijleo.2016.05.071
  2. D. Ngo, G. D. Lee, and B. S. Kang, "Improved color attenuation prior for single-image haze removal," Appl. Sci., vol.9, no.19, pp.4011, 2019. DOI: 10.3390/app9194011
  3. Q. Zhu, J. Mai, and L. Shao, "A Fast Single Image Haze Removal Algorithm Using Color Attenuation Prior," IEEE Trans. Image Process., vol.24, no.11, pp.3522-3533, 2015. DOI: 10.1109/TIP.2015.2446191
  4. K. He, J. Sun, and X. Tang, "Single Image Haze Removal Using Dark Channel Prior," IEEE Trans. Pattern Anal. Mach. Intell., vol.33, no.12, pp.2341-2353, 2011. DOI: 10.1109/TPAMI.2010.168
  5. J. P. Tarel and N. Hautiere, "Fast visivility restoration from a single color or gray level image," in Proc. of the 2009 IEEE International Conference on Computer Vision, pp.2201-2208, 2009. DOI: 10.1109/ICCV.2009.5459251
  6. D. Ngo and B. S. Kang, "Improving Performance of Machine Learning-based Haze Removal Algorithms with Enhanced Training Database," J. Ins. Korean. Electr. Electron. Eng., vol.22, no.4, pp.948-952, 2018. DOI: 10.7471/ikeee.2018.22.4.948
  7. J. H. Kim, J. Y. Sim, and C. S. Kim, "Single image dehazing based on contrast enhancement," in Proc. of the 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.1273-1276, 2011. DOI: 10.1109/ICASSP.2011.5946643
  8. B. Cai, X. Xu, K. Jia, C. Qing, D. Tao, "Dehaze-Net: An End-to-End System for Single Image Haze Removal," IEEE Trans. Image Process., vol.25, no.11, pp.5187-5198, 2016. DOI: 10.1109/TIP.2016.2598681

Cited by

  1. CIE1931 색좌표계 변환의 최적화된 하드웨어 구현을 통한 색상 보정 vol.25, pp.1, 2021, https://doi.org/10.7471/ikeee.2021.25.1.10