Single image dehazing via reliability guided fusion

Riaz, Irfan; Teng, Yunlong; Rehman, Yawar; Shin, Hyunchul

doi:10.1016/j.jvcir.2016.06.011

Cited by 35 publications

(12 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…He et al [33] generalized dark-object subtraction method and proposed the Dark Channel Prior (DCP). DCP method produced excellent results, hence much work after that [22,51,52,64,21,45,33,38] focused on its drawbacks for improvement. Carr [13] used piece-wise planar-geometry with DCP and energy minimization framework.…”

Section: Related Workmentioning

confidence: 99%

“…In [16] Dharna et al proposed color correction on color cast images by performing weighted least squares filtering on DCP. The works of [51,52,64] estimated the transmission for bright regions by considering the values of neighborhood values. The technique proposed by Riaz et al [52] increase patch size for minimizing probability of DCP failure and halos.…”

Section: Related Workmentioning

confidence: 99%

“…The works of [51,52,64] estimated the transmission for bright regions by considering the values of neighborhood values. The technique proposed by Riaz et al [52] increase patch size for minimizing probability of DCP failure and halos. Yu et al [64] suggested block-to-pixel interpolation method to accelerate transmission estimation process in DCP.…”

Section: Related Workmentioning

confidence: 99%

“…Figures 6-8 present images with large scene depth and sharp depth variations. Such images are challenging as most de-hazing algorithms [51,52,27,19,64,20,49,11,50] assume that transmission remains constant within a small patch of neighborhood pixels. This assumption becomes invalid at depth discontinuity of object boundaries, which causes halo artifacts to appear around objects in the de-hazed image.…”

Section: Qualitative Analysismentioning

confidence: 99%

“…Here, I(x) is an RGB color hazy image, J(x) represents scene radiance, t(x) is visibility map or transmission of an image and A is airlight. Illumination is commonly assumed to be uniform across a scene [41,54,44,26,22,51,52], hence airlight A becomes a three-dimensional constant vector, which is often assumed to be the brightest point in the haziest region of the image [9]. Transmission t(x) depicts the proportion of scene radiance J(x) approaching the camera directly without being scattered.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Visibility Restoration Using Generalized Haze-Lines

Riaz¹,

Anwar

Riaz³

et al. 2021

ITC

Self Cite

View full text Add to dashboard Cite

Haze reduces the perceived scene radiance and limits the visibility in outdoor images. The visibility is different for each scene point and is proportional to haze thickness, and distance from the camera. Transmission map represents percentage of scene radiance captured by the camera and is unknown for every pixel. This work generalizes the concept of haze-lines, and presents an algorithm to estimate transmission map and restore scene radiance accurately. The proposed technique depends on the perception that the colors of haze-free natural images can be well approximated by a set of distinct colors and their shades (natural color-palette) that can be learned beforehand. In presence of haze, the pixels forming a cluster in haze-free image, make a line (haze-line) in RGB color space. The two endpoints of this haze-line are the haze-free color and the airlight. We propose that these haze-lines can be generalized, with one end as learned color-palette of natural images and the other as airlight. Hence the scene radiance end can be made independent of underlying image. The algorithm recovers the transmission map, by determining membership of each pixel to a given haze-line and finding how far-off it is from its learned color-palette. The algorithm is linear to the size of image, and requires just a collection of haze-free natural images for training. The results obtained on a diverse range of images demonstrate the efficiency of proposed algorithm.

show abstract