On the Duality Between Retinex and Image Dehazing

@article{Galdran2018OnTD,
  title={On the Duality Between Retinex and Image Dehazing},
  author={Adrian Galdran and Aitor Alvarez-Gila and Alessandro Bria and Javier Vazquez-Corral and Marcelo Bertalm{\'i}o},
  journal={2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2018},
  pages={8212-8221}
}
Image dehazing deals with the removal of undesired loss of visibility in outdoor images due to the presence of fog. Retinex is a color vision model mimicking the ability of the Human Visual System to robustly discount varying illuminations when observing a scene under different spectral lighting conditions. Retinex has been widely explored in the computer vision literature for image enhancement and other related tasks. While these two problems are apparently unrelated, the goal of this work is… 

Figures and Tables from this paper

Image dehazing by artificial multiple-exposure image fusion
Physical-based optimization for non-physical image dehazing methods
TLDR
This paper proposes a post-processing technique to alleviate the handicap of image dehazing by enforcing the original method to be consistent with a popular physical model for image formation under haze, and its results improve upon those of the original methods qualitatively and according to several metrics.
Low Light Image Enhancement Algorithm Based on Retinex and Dehazing Model
TLDR
Compared with the existing mainstream algorithms, the proposed algorithm has the advantages of avoiding over exposure, rich texture details, low noise and high color recovery.
A Unified Variational Model for Single Image Dehazing
TLDR
A simple but effective unified variational model for single image dehazing is presented based on the total variation regularization, which can simultaneously obtain the accurate transmission map and the recovered scene radiance by integrating the transmission estimation stage and the image recovery stage into the unified Variational model.
Variational Single Image Dehazing for Enhanced Visualization
TLDR
A variational model is proposed in the Y channel of the YUV color space by combining the reformulation of the haze model and the two effective priors, which can unveil the details and recover vivid color information, which is competitive among many existing dehazing algorithms.
Physical-based optimization for non-physical image dehazing methods.
TLDR
This paper proposes a post-processing technique to alleviate the handicap of image dehazing by enforcing the original method to be consistent with a popular physical model for image formation under haze, which improves upon those of the original methods qualitatively and according to several metrics.
Deep Retinex Network for Single Image Dehazing
TLDR
A deep retinex dehazing network (RDN) to jointly estimate the residual illumination map and the haze-free image and can avoid the errors associated with the simplified scattering model and provide better generalization ability with no dependence on prior information.
A General-Purpose Dehazing Algorithm based on Local Contrast Enhancement Approaches
TLDR
This document presents a dehazing method which is suitable for several local contrast adjustment algorithms, and shows promising results compared to state-of-the-art algorithms based not only on a visual assessment but also on objective criteria.
Evaluation of image dehazing techniques based on a realistic benchmark
TLDR
A new realistic dehazing dataset that consists of 22 pairs of hazy images and haze-free images (ground truth) providing also a quantitative validation of the existing dehaing methods using the PSNR, SSIM and CIEDE2000 objective measures.
RESUSCITATED VARIATION MODEL FOR IMAGE DEHAZING
TLDR
An effective method for fog removal is presented, which uses retrieved correlated fog-free images as external information and simultaneously integrates the transmission estimation stage and image retrieval stage into a unified prediction model to obtain accurate transmission map and retrieved visual illumination.
...
...

References

SHOWING 1-10 OF 55 REFERENCES
Non-local Image Dehazing
TLDR
This work proposes an algorithm, linear in the size of the image, deterministic and requires no training, that performs well on a wide variety of images and is competitive with other state-of-the-art methods on the single image dehazing problem.
Enhanced Variational Image Dehazing
TLDR
Experimental results show that the proposed enhanced variational image dehazing (EVID) method outperforms other state-of-the-art methods both qualitatively and quantitatively.
Removing Shadows From Images using Retinex
TLDR
The Retinex Theory can be extended to perform yet another image processing task: that of removing shadows from images by a simple modification to the original, path based retinex computation such that it incorporates information about the location of shadow edges in an image.
Single Image Haze Removal Using Dark Channel Prior
TLDR
This thesis develops an effective but very simple prior, called the dark channel prior, to remove haze from a single image, and thus solves the ambiguity of the problem.
A Variational Framework for Retinex
TLDR
A variational model for the Retinex problem that unifies previous methods and shows that the illumination estimation problem can be formulated as a Quadratic Programming optimization problem.
Fast visibility restoration from a single color or gray level image
TLDR
A novel algorithm and variants for visibility restoration from a single image which allows visibility restoration to be applied for the first time within real-time processing applications such as sign, lane-marking and obstacle detection from an in-vehicle camera.
Automatic recovery of the atmospheric light in hazy images
TLDR
This paper presents a new automatic method for recovering the atmospheric light vector in hazy scenes given a single input image by exploiting a global regularity which is observed in haazy images where the intensity level of the brightest pixels is approximately independent of their transmission value.
Dehazing Using Color-Lines
TLDR
A new method for single-image dehazing that relies on a generic regularity in natural images where pixels of small image patches typically exhibit a 1D distribution in RGB color space, known as color-lines is described.
Stress for dehazing
TLDR
A new image enhancement framework named STRESS (Kol°as et al., JIST, 55(4), 2011) which has some specificities similar to the Retinex algorithm is made and compared with some of state-of-the-art dehazing algorithms and STRESS model turned out to be better than almost any state of theart algorithms in terms of the quality of the images output.
LIME: Low-Light Image Enhancement via Illumination Map Estimation
TLDR
Experiments on a number of challenging low-light images are present to reveal the efficacy of the proposed LIME and show its superiority over several state-of-the-arts in terms of enhancement quality and efficiency.
...
...