Non-local Image Dehazing

Berman, Dana; Treibitz, Tali; Avidan, Shai

doi:10.1109/cvpr.2016.185

Cited by 1,231 publications

(1,171 citation statements)

References 18 publications

Supporting

Mentioning

1,084

Contrasting

Unclassified

Order By: Relevance

“…To handle the block artifact caused by the patch based estimation, guided filtering [48] is used to refine the transmission map. Berman et al 2016 [18] proposes an algorithm based on a new, non-local prior. This is a departure from existing methods (e.g.…”

Section: E Single-image Methodsmentioning

confidence: 99%

“…that use patch based transmission estimation. The algorithm by [18] relies on the assumption that colors of a haze-free image are well approximated by a few hundred distinct colors, that form tight clusters in RGB space and pixels in a cluster are often nonlocal (spread in the whole images). The presence of haze will elongate the shape of each cluster to a line in color space as the pixels may be affected by different transmission coefficients due to their different distances to the camera.…”

Section: E Single-image Methodsmentioning

confidence: 99%

“…Figure 8 shows the performance of each method in term of SSIM. It is observed that again latest methods Tang 14 [43], Cai 16 [46], and Berman 16 [18] generally performed better than others. He 09 [17] also performs very well, especially in heavier haze levels.…”

Section: B Evaluation On Our Datasetmentioning

confidence: 98%

“…Accordingly, given a pixel and the two chromaticity values, we can immediately calculate the magnitude of the airlight, which consequently gives us the absolute depth map by employing Eq. (18). In this algorithm, the chromaticity of the direct transmission is assumed to be given from a clear day image of exactly the same scene, and the airlight chromaticity is computed from a known atmospheric light.…”

Section: A Early Work In Depth Estimationmentioning

confidence: 99%

See 3 more Smart Citations

Haze visibility enhancement: A Survey and quantitative benchmarking

Liu

You

Brown

et al. 2017

Computer Vision and Image Understanding

149

View full text Add to dashboard Cite

Abstract-This paper provides a comprehensive survey of methods dealing with visibility enhancement of images taken in hazy or foggy scenes. The survey begins with discussing the optical models of atmospheric scattering media and image formation. This is followed by a survey of existing methods, which are grouped to multiple image methods, polarizing filters based methods, methods with known depth, and single-image methods. We also provide a benchmark of a number of well known single-image methods, based on a recent dataset provided by Fattal [1] and our newly generated scattering media dataset that contains ground truth images for quantitative evaluation. To our knowledge, this is the first benchmark using numerical metrics to evaluate dehazing techniques. This benchmark allows us to objectively compare the results of existing methods and to better identify the strengths and limitations of each method.

show abstract

Section: E Single-image Methodsmentioning

confidence: 99%

Section: E Single-image Methodsmentioning

confidence: 99%

Section: B Evaluation On Our Datasetmentioning

confidence: 98%

Section: A Early Work In Depth Estimationmentioning

confidence: 99%

See 2 more Smart Citations

Haze visibility enhancement: A Survey and quantitative benchmarking

Liu

You

Brown

et al. 2017

Computer Vision and Image Understanding

149

View full text Add to dashboard Cite

show abstract

“…The pixels with the same diffuse chromaticity (x) make (5) a three-dimensional line, and (5) is a lines set for an image with various diffuse chromaticities. In recent literature, the authors of [26] and [27] also use color lines for haze removal. However, their color lines model the medium transmission that describes proportion between ambient light and scene radiance, while our color lines model the surface reflection properties of an object, namely the relation between illumination chromaticity and diffuse chromaticity.…”

Section: Global Color-lines Constraint a Global Color-lines Conmentioning

confidence: 99%

Specular Reflection Separation With Color-Lines Constraint

Ren

Tian

Tang

2017

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Abstract-According to dichromatic reflection model, the previous methods of specular reflection separation in image processing often separate specular reflection from a single image using patch-based priors. Due to lack of global information, these methods often cannot completely separate the specular component of an image and are incline to degrade image textures. In this paper, we derive a global color-lines constraint from dichromatic reflection model to effectively recover specular and diffuse reflection. Our key observation is from that each image pixel lies along a color line in normalized RGB space and the different color lines representing distinct diffuse chromaticities intersect at one point, namely, the illumination chromaticity. For pixels along the same color line, they spread over the entire image and their distances to the illumination chromaticity reflect the amount of specular reflection components. With global (non-local) information from these color lines, our method can effectively separate specular and diffuse reflection components in a pixelwise way for a single image, and it is suitable for realtime applications. Our experimental results on synthetic and real images show that our method performs better than the state-ofthe-art methods to separate specular reflection.

show abstract

Color corrected single scale retinex based haze removal and color correction for underwater images

Bindhu

Maheswari

2020

Color Research & Application

View full text Add to dashboard Cite

Underwater images are affected by scattering of dust particles leading to hazy images. The difference in absorption of light at different wavelength leads to colorcast. Restoring underwater images from haze and colorcast degradation is a challenging task. This article proposes color corrected single scale retinex algorithm for the restoration of hazy and colorcast underwater images. Further, the illumination of the underwater image is enhanced through illumination enhancer. In order to correct colorcast, the underwater images intensity shifters are proposed. Blur, natural image quality evaluator, blind/ referenceless image spatial quality evaluator, entropy, peak signal to noise ratio are the metrics used to evaluate the performance of the proposed restoration technique and compared the output with the existing techniques. Compared to the existing techniques the proposed algorithm prominently removes haze and colorcast present in underwater images. The study excludes retinexwith-reset models from consideration.

show abstract

Non-local Image Dehazing

Cited by 1,231 publications

References 18 publications

Haze visibility enhancement: A Survey and quantitative benchmarking

Haze visibility enhancement: A Survey and quantitative benchmarking

Specular Reflection Separation With Color-Lines Constraint

Color corrected single scale retinex based haze removal and color correction for underwater images

Contact Info

Product

Resources

About