Haze-free Image Research Articles

A Single Image Dehazing Technique Using the Dual Transmission Maps Strategy and Gradient-Domain Guided Image Filtering

In this paper, a single image dehazing technique using dual transmission maps strategy and gradient-domain guided image filtering is presented. A new strategy is adopted to compute the dual transmission maps using the dark channel and atmospheric light. Further, the transmission maps are refined to remove any remaining ill effects using the gradient-domain-guided filter. Finally, using the dark channel, atmospheric light, and refined transmission map, the haze-free image is obtained. The dual transmission maps strategy not only removes halo artifacts and reduces the saturation but also ensures the natural appearance in the recovered images. Furthermore, the proposed scheme is evaluated using a wide range of images and compared with state-of-the-art schemes. The comparison shows the superiority of the proposed technique in terms of recovering haze-free images.

RefineDNet: A Weakly Supervised Refinement Framework for Single Image Dehazing

Haze-free images are the prerequisites of many vision systems and algorithms, and thus single image dehazing is of paramount importance in computer vision. In this field, prior-based methods have achieved initial success. However, they often introduce annoying artifacts to outputs because their priors can hardly fit all situations. By contrast, learning-based methods can generate more natural results. Nonetheless, due to the lack of paired foggy and clear outdoor images of the same scenes as training samples, their haze removal abilities are limited. In this work, we attempt to merge the merits of prior-based and learning-based approaches by dividing the dehazing task into two sub-tasks, i.e., visibility restoration and realness improvement. Specifically, we propose a two-stage weakly supervised dehazing framework, RefineDNet. In the first stage, RefineDNet adopts the dark channel prior to restore visibility. Then, in the second stage, it refines preliminary dehazing results of the first stage to improve realness via adversarial learning with unpaired foggy and clear images. To get more qualified results, we also propose an effective perceptual fusion strategy to blend different dehazing outputs. Extensive experiments corroborate that RefineDNet with the perceptual fusion has an outstanding haze removal capability and can also produce visually pleasing results. Even implemented with basic backbone networks, RefineDNet can outperform supervised dehazing approaches as well as other state-of-the-art methods on indoor and outdoor datasets. To make our results reproducible, relevant code and data are available at https://github.com/xiaofeng94/RefineDNet-for-dehazing.

Single Image Dehazing Using Wavelet-Based Haze-Lines and Denoising

Haze reduces the contrast of an image and causes the loss in colors, which has a negative effect on the subsequent object detection; therefore, single image dehazing is a challenging visual task. In addition, defects exist in previous existing dehazing approaches: Pixel-based dehazing approaches are likely to result in insufficient information to estimate the transmission, whereas patch-based ones are prone to generate shadows. They both also tend to induce color deviations. Therefore, this study proposes a novel method based on multi-scale wavelet and non-local dehazing. A hazy image is first decomposed into a low-frequency and three high-frequency sub-images by wavelet transform. Non-local dehazing and wavelet denoising are then employed on the low-frequency and high-frequency sub-images to remove the haze and noise, respectively. Finally, a haze-free image is obtained from the reconstruction of sub-images. The proposed method focuses on the dehazing and denoising on the low-frequency and high-frequency images respectively, through which the details on the image can be well preserved. Experimental results indicate that the proposed method is superior to the state-of-the-art approaches in both quantitative and qualitative evaluation on the synthetic and real-world image datasets.

Single Image Dehazing of Road Scenes Using Spatially Adaptive Atmospheric Point Spread Function

Image haze removal is essential in autonomous driving as the outdoor images captured during unfavorable weather conditions, such as haze or snow, are affected by poor visibility. Much research has been done to overcome image degradation such as low contrast and faded color due to haze. However, in the traditional model, a phenomenon is neglected that several particles simultaneously involved in light acquisition. To address this problem, we propose a novel single image dehazing method based on the spatially adaptive atmospheric point spread function (APSF). We developed a module that estimates the APSF to overcome the limitations of the spatially invariant APSF which used in existing dehazing algorithms. The key factor in the estimation is that road scenes with haze have different statistical characteristic from common hazy images in color and resolution. Furthermore, the APSF on the traffic signs or lights is estimated by generating superpixels to prevent halo artifacts around the sharp edges of the images. We adopted the total variation model as a regularization functional to reduce halo and unnatural artifacts that may occur during deconvolution. The haze-free images from the proposed method tested whether the proposed method can enhance the performance of vision algorithms for autonomous driving. The experimental results demonstrate that the proposed method outperforms state-of-the-art image dehazing methods enhancing the performance of the vision algorithms. Moreover, additional experiments demonstrated the effectiveness of the proposed method for quantitative and qualitative comparison with the state-of-the-art algorithms.

Single-Image Dehazing Using Extreme Reflectance Channel Prior

Image dehazing algorithms based on dark channel prior principle have achieved good results for most scenes. However, the popular dark channel prior tends to underestimate transmissions of bright areas or objects, such as the skies, white areas and self-luminous bodies, which may cause color distortions during dehazing. A complementary prior called the extreme reflectance channel prior (ERC), which combines the dark channel prior with the bright channel prior, is proposed to estimate the transmission map. The extreme reflectance channel is the union of dark and bright channel's pixels which satisfy the corresponding channel. Based on the scattering analysis results that the intensities of pixels in ERC are often close to 0 or 1 for the natural haze-free images or close to global atmospheric light if hazes occur in the air, the pixels in a hazy image can be recovered according to ERC to calculate the transmission map and then solve the haze imaging mode. Experiments show that ERC method outperforms state-of-the-art methods in PSNR, SSIM and visual perception effects.

Pixel-Wise Wasserstein Autoencoder for Highly Generative Dehazing.

We propose a highly generative dehazing method based on pixel-wise Wasserstein autoencoders. In contrast to existing dehazing methods based on generative adversarial networks, our method can produce a variety of dehazed images with different styles. It significantly improves the dehazing accuracy via pixel-wise matching from hazy to dehazed images through 2-dimensional latent tensors of the Wasserstein autoencoder. In addition, we present an advanced feature fusion technique to deliver rich information to the latent space. For style transfer, we introduce a mapping function that transforms existing latent spaces to new ones. Thus, our method can produce highly generative haze-free images with various tones, illuminations, and moods, which induces several interesting applications, including low-light enhancement, daytime dehazing, nighttime dehazing, and underwater image enhancement. Experimental results demonstrate that our method quantitatively outperforms existing state-of-the-art methods for synthetic and real-world datasets, and simultaneously generates highly generative haze-free images, which are qualitatively diverse.

Multi-Scale Single Image Dehazing Using Laplacian and Gaussian Pyramids.

Model-based single image dehazing was widely studied due to its extensive applications. Ambiguity between object radiance and haze and noise amplification in sky regions are two inherent problems of model-based single image dehazing. In this paper, a dark direct attenuation prior (DDAP) is proposed to address the former problem. A novel haze line averaging is proposed to reduce the morphological artifacts caused by the DDAP which enables a weighted guided image filter with a smaller radius to further reduce the morphological artifacts while preserve the fine structure in the image. A multi-scale dehazing algorithm is then proposed to address the latter problem by adopting Laplacian and Gaussian pyramids to decompose the hazy image into different levels and applying different haze removal and noise reduction approaches to restore the scene radiance at the different levels. The resultant pyramid is collapsed to restore a haze-free image. Experiment results demonstrate that the proposed algorithm outperforms state-of-the-art dehazing algorithms.

Deep Dehazing Network With Latent Ensembling Architecture and Adversarial Learning.

Most existing dehazing algorithms recover haze-free image by solving the hazy imaging model using estimated transmission map and global atmospheric light. However, inaccurate estimation of these variables and the strong assumptions of imaging model result in unrealistic dehazing results. In this paper, we use the adversarial game between a pair of neural networks to accomplish end-to-end photo-realistic dehazing. To avoid uniform contrast enhancement, the generator learns to simultaneously restore haze-free image and capture the non-uniformity of haze. The modules for the two tasks are assembled in sequential and parallel manners to enable information sharing at different levels, and the architecture of the generator implicitly forms an ensemble of dehazing models that allows for feature selection. A multi-scale discriminator competes with the generator by learning to detect dehazing artifacts and the inconsistency between dehazed image and the spatial variation of haze. Unlike existing works that penalize dehazing artifacts via hand-crafted loss, the proposed algorithm uses the identity mapping in the space of clear-scene images to regularize data-driven dehazing. The proposed work also addresses the adaptability of data-driven dehazing to high-level computer vision task. We propose a task-driven training strategy that can optimize the object detection performance on dehazed images without updating the parameters of object detector. Performance of the proposed algorithm is assessed on the RESIDE, I-Haze, and O-Haze benchmarks. The comparison with ten state-of-the-art algorithms shows that the proposed work is the best performer in most competitions.

Deep Retinex Network for Single Image Dehazing.

In this paper, we propose a retinex-based decomposition model for a hazy image and a novel end-to-end image dehazing network. In the model, the illumination of the hazy image is decomposed into natural illumination for the haze-free image and residual illumination caused by haze. Based on this model, we design a deep retinex dehazing network (RDN) to jointly estimate the residual illumination map and the haze-free image. Our RDN consists of a multiscale residual dense network for estimating the residual illumination map and a U-Net with channel and spatial attention mechanisms for image dehazing. The multiscale residual dense network can simultaneously capture global contextual information from small-scale receptive fields and local detailed information from large-scale receptive fields to precisely estimate the residual illumination map caused by haze. In the dehazing U-Net, we apply the channel and spatial attention mechanisms in the skip connection of the U-Net to achieve a trade-off between overdehazing and underdehazing by automatically adjusting the channel-wise and pixel-wise attention weights. Compared with scattering model-based networks, fully data-driven networks, and prior-based dehazing methods, our RDN can avoid the errors associated with the simplified scattering model and provide better generalization ability with no dependence on prior information. Extensive experiments show the superiority of the RDN to various state-of-the-art methods.

URNet: A U-Net based residual network for image dehazing

Low visibility in hazy weather causes the loss of image details in digital images captured by some imaging devices such as monitors. This paper proposes an end-to-end U-Net based residual network (URNet) to improve the visibility of hazy images. The encoder module of URNet uses hybrid convolution combining standard convolution with dilated convolution to expand the receptive field for extracting image features with more details. The URNet embeds several building blocks of ResNet into the junction between the encoder module and the decoder module. This prevents network performance degradation due to the vanishing gradient. After considering large absolute difference on image saturation and value components between hazy images and haze-free images in the HSV color space, the URNet defines a new loss function to better guide the network training. Experimental results on synthetic hazy images and real hazy images show that the URNet significantly improves the image dehazing effect compared to the state-of-the-art methods.

Image dehazing method via a cycle generative adversarial network

Aiming at the problems of colour distortion and residual dehazing in the existing image dehazing methods when processing outdoor images, a method of image dehazing based on the cycle generative adversarial network is proposed. Taking the cycle generative adversarial network as the overall framework of the model, firstly, the neural network is trained to obtain the mapping relationship between haze images and haze-free images. Secondly, to accelerate the convergence speed of network training, using residual structure to improve network stability and reduce parameters. Then, a new loss function is proposed, fusing the Wasserstein distance into adversarial loss and cycle consistency loss, which reduces the deviation between the generated dehaze image and the real haze-free image, and alleviating the problems of colour distortion and haze removal residue. Finally, use the optimised bounded ReLU (BReLU) activation function instead of the original activation function to improve the transmittance reflecting the haze depth information. The experimental results demonstrated that, compared with the comparison method, the proposed method improves the peak signal-to-noise ratio, structure similarity, information entropy, and average gradient, and has an achieved better performance in dehazing.

Single image haze removal using variable fog-weight

In this paper, a new Image Haze Removal technique is proposed using variable fog-weight. The objective of the proposed technique is to improve the visibility, to remove the impact of weather factors and make image more usable. To achieve this, it is required to estimate the transmission map as well as atmospheric light by using Dark Channel Prior. The main contribution in this work, is to vary the fog-weight during transmission estimation. It is varied according to the density of haze in the input image. If the density of the haze in the input image is low, then lower fog-weight value will be applied and if the haze density in the input image is higher, than higher fog-weight value will be applied, according to the haze concentration. A guided filter is used as a refiner of transmission map to avoid the halo artifacts at the boundary of the edge. It provides smoothness to the image and build-up the visibility of the image, so that we restore a superior haze free image. At last, the performance of the proposed method is compared with some of the existing methods in terms of visibility, computational time and avoidance of halo effect.

CGGAN: a context‐guided generative adversarial network for single image dehazing

Image haze removal is highly desired for the application of computer vision. This study proposes a novel context-guided generative adversarial network (CGGAN) for single image dehazing. Of which, a novel new encoder–decoder is employed as the generator. In addition, it consists of a feature-extraction net, a context-extraction net, and a fusion net in sequence. The feature-extraction net acts as an encoder, and is used for extracting haze features. The content-extraction net is a multi-scale parallel pyramid decoder and is used for extracting the deep features of the encoder and generating coarse dehazing image. The fusion net is a decoder and is used for obtaining the final haze-free image. In order to get better dehazing results, multi-scale information obtained during the decoding process of the context extraction decoder is used for guiding the fusion decoder. By introducing an extra coarse decoder to the original encoder–decoder, the CGGAN can make better use of the deep feature information extracted by the encoder. To ensure that the proposed CGGAN works effectively for different haze scenarios, different loss functions are employed for the two decoders. Experiments results show the advantage and the effectiveness of the proposed CGGAN, evidential improvements over existing state-of-the-art methods are obtained.

An End-to-End Network for Image De-Hazing and Beyond

Degradation in the qualityof images that are captured in the hazy environment is mainlydue to 1) different weather conditions and 2) the attenuation in reflected light. These factors introduce a severe color distortion and low visibility in the captured images. To tackle these problems, we propose an end-to-end trainable image de-hazing network named as LIGHT-Net. The proposed LIGHT-Net comprises of color constancy module and haze reduction module. Among these, the color constancy module removes the color cast added in hazy image due to the weather condition. Whereas, the proposed haze reduction module, which is build using an inception-residual block, is aimed to reduce the effect of haze as well as to improve the visibility in the hazy image. Unlike traditional feature concatenation, in the haze reduction module, we propose a dense feature sharing to effectively share the features learned at initial layers across the network. In general, a major hurdle to train a convolution neural network for haze removal task is the unavailability of large-scale real-world hazy, and corresponding haze-free image (i.e. paired data). Thus, we make use of an unpaired training approach to train the proposed LIGHT-Net for image de-hazing. Extensive analysis has been carried out to validate the necessity and impact of each sub-block of the proposed LIGHT-Net. A large set of real-world hazy images captured in different weather conditions are considered to validate the proposed approach for image de-hazing. Also, the benchmark synthetic hazy image database is considered for a quantitative analysis of the proposed LIGHTNet for image de-hazing. Further, we have shown the usefulness of the proposed LIGHT-Net for underwater image enhancement. Experiments show that the proposed LIGHT-Net outperforms the other existing approaches for both image de-hazing as well as underwater image enhancement.

Single image dehazing by dark channel prior and luminance adjustment

ABSTRACT This paper presents a new and simple algorithm to eliminate haze from a single image using YIQ colour space and Dark Channel Prior (DCP). The suggested method consists of two parts. The first one is enhancement by adjusting contrast and lightness using contrast-limited adaptive histogram equalization and gamma transform. The second part involves applying DCP on a hazy image using the RGB colour model to achieve a haze-free image. The resulting image is then converted into YIQ, and the chrominance components (I and Q) are extracted and combined with the enhanced luminance component (Y) to achieve an enhancement haze-free image. The suggested method is applied to many outdoor images, and the quality of the resulting images is compared with that of images from other methods, by using several measures of quality. Analysing the results explain that the suggested method has a higher quality value than other methods.

ICycleGAN: Single image dehazing based on iterative dehazing model and CycleGAN

The current competitive approaches to restoring haze-free images are mainly based on physical models and learning methods. Maintaining detail information of the image while thoroughly removing fog is a challenging task in single-image dehazing. In this paper, by embedding an iterative dehazing model into the generative process of the Cycle-Consistent Adversarial Network (CycleGAN), we propose a model named ICycleGAN that maintains the defogging thoroughness of the learning-based dehazing method while retaining the good fidelity of the physical model-based dehazing method. Moreover, the proposed ICycleGAN does not require pairs of hazy and relative haze-free images for training. In addition, we develop a detail information-consistency loss that preserves more textural details and color information; this loss is obtained based on the physical features of the hazy image. To recover the high-resolution images, we enlarge the generated images using rational fractal interpolation, which restores fine textures and sharp edges. Extensive comparison results show that the proposed method produces high-quality clear images that are both quantitatively and qualitatively competitive with other state-of-the-art methods.

Unsupervised Dark-Channel Attention-Guided CycleGAN for Single-Image Dehazing.

In this paper, we propose a new unsupervised attention-based cycle generative adversarial network to solve the problem of single-image dehazing. The proposed method adds an attention mechanism that can dehaze different areas on the basis of the previous generative adversarial network (GAN) dehazing method. This mechanism not only avoids the need to change the haze-free area due to the overall style migration of traditional GANs, but also pays attention to the different degrees of haze concentrations that need to be changed, while retaining the details of the original image. To more accurately and quickly label the concentrations and areas of haze, we innovatively use training-enhanced dark channels as attention maps, combining the advantages of prior algorithms and deep learning. The proposed method does not require paired datasets, and it can adequately generate high-resolution images. Experiments demonstrate that our algorithm is superior to previous algorithms in various scenarios. The proposed algorithm can effectively process very hazy images, misty images, and haze-free images, which is of great significance for dehazing in complex scenes.

Single image dehazing based on single pixel energy minimization

The common dehazing algorithms always assume that the transmission values of all the pixels in an image block are the same (local consistency assumption). However, it is easy to appear “halo” for image regions where the depth changes obviously. In this paper, we calculate the transmission of each pixel separately without the local consistency assumption. First, we initialize a random transmission value for each pixel in the whole image. Then, we optimize the transmission values through several iterations by minimizing an energy function, which contains the data term and penalty term. In each iteration, we take two procedures of propagation and random search to optimize transmission values. Finally, we use the optimized transmission and the estimated atmospheric light to calculate the haze-free image. Comparison experiments show that our algorithm can remove haze effectively, and obtain the best performance.

Single Image Dehazing by Predicting Atmospheric Scattering Parameters

In this work we propose a method for single image dehazing that exploits a physical model to recover the haze-free image by estimating the atmospheric scattering parameters. Cycle consistency is used to further improve the reconstruction quality of local structures and objects in the scene as well. Experimental results on four real and synthetic hazy image datasets show the effectiveness of the proposed method in terms of two commonly used full-reference image quality metrics.

Enhancement of Hazy Images Using Atmospheric Light Estimation Technique

Images captured in degraded weather conditions often suffer from bad visibility. Pre-existing haze removal methods, the ones that are effective are computationally complex too. In common de-hazing approaches, estimation of atmospheric light is not achieved properly as a consequence, haze is not removed significantly from the sky region. In this paper, an efficient method of haze removal from a single image is introduced. To restore haze-free images comprising of both sky as well as nonsky regions, we developed a linear model to predict atmospheric light and estimated the transmission map using the dark channel prior followed by an application of a guided filter for quick refinement. Several experiments were conducted on a large variety of images, both reference and nonreference, where the proposed image de-hazing algorithm outperforms most of the prevalent algorithms in terms of perceptual visibility of the scene and computational efficiency. The proposed method has been empirically measured through quantitative and qualitative evaluations while retaining structure, edges, and improved color.