Search

Haze is a natural phenomenon that negatively affects image clarity and quality, posing challenges across various image-related applications. Traditional dehazing models often suffer from overfitting when trained on synthetic hazy-clean image pairs, which do not generalize well to real-world hazy conditions. To tackle this, recent methodologies have explored training models on unpaired data, better reflecting the variability encountered in natural scenes. This dual capability of CycleGAN is particularly beneficial for overcoming the overfitting issues associated with synthetic datasets. By incorporating CycleGAN into our DehazeDNet framework, we ensure that our dehazing model not only translates images effectively but also respects the physical characteristics of haze. Inspired by the D4 model, our approach includes a Depth Evaluation Block to estimate scene depth from images. Since haze density often correlates with scene depth, this depth information is crucial for accurate haze modeling. We utilize the U-Net architecture for the Depth Evaluation Block due to its proven efficiency in image-to-image translation tasks. To preserve the accuracy of the dehazed images, we incorporate an identity loss function into our model. Identity loss ensures that the dehazed output retains the essential characteristics of the input image. Our results demonstrate an increase in SSIM and PSNR compared to other unsupervised dehazing models, highlighting the efficiency of our method in maintaining image quality and details while removing haze. [ABSTRACT FROM AUTHOR]

This paper proposes an innovative single image dehazing framework, termed Regional Saturation-Value Translation (RSVT), to address the color distortion problems commonly encountered in bright regions by conventional dehazing approaches. The proposed RSVT framework is developed based on two key insights derived from the HSV color space: first, the hue component shows negligible variation between corresponding hazy and haze-free points; and second, in the 2D saturation-value coordinate system, the majority of lines connecting hazy-clean point pairs tend to converge near the atmospheric light coordinates. Consequently, haze removal can be achieved through appropriate translations within the saturation-value coordinates. Additionally, a robust soft segmentation method that employs a morphological min-max channel is integrated into the framework. By combining the soft segmentation mask with the RSVT prior, a comprehensive single image dehazing framework is established. Experimental evaluations across various datasets demonstrate that the proposed approach effectively mitigates color distortion and successfully restores visually appealing images. Moreover, a case study involving actual flight test demonstrates the feasibility and effectiveness of the proposed approach in real-world scenarios. The code is available at https://github.com/tranleanh/rsvt.

This paper proposes an image dehazing prior, called Regional Saturation-Value Translation (RSVT), in order to address the color distortion issues produced by prevailing prior-based dehazing methods when processing hazy images with large sky regions. The proposed RSVT prior is derived from statistical analyses of the correlation between hazy points and respective haze-free points in the HSV color space. The prior is based upon two key observations in the sky areas. First, the difference in terms of hue for a pair of hazy and haze-free points is very small, raising an assumption that the variability of pixel values caused by haze mostly occurs in the saturation and value spaces. This leads to the second observation that, in the 2D saturation-value coordinate system, almost all the lines passing through corresponding pairs of hazy-clean points, termed S-V lines, are likely to intersect around the airlight coordinates. A hybrid refined dark channel is introduced in order to decompose the input hazy image into sky and non-sky regions and to estimate the global atmospheric light. Combining the prior with the hybrid refined dark channel, a novel single image dehazing framework is proposed. Haze removal is performed separately for the sky and non-sky regions by adopting the proposed RSVT prior and Koschmieder's law, respectively. The experimental results have shown that the proposed dehazing method can restore visually compelling sky color and effectively handle the color distortion issues associated with large sky regions. [ABSTRACT FROM AUTHOR]

The existing remote sensing (RS) image dehazing methods based on deep learning have sought help from the convolutional frameworks. Nevertheless, the inherent limitations of convolution, i.e., local receptive fields and independent input elements, curtail the network from learning the long-range dependencies and nonuniform distributions. To this end, we design an effective RS image dehazing Transformer architecture, denoted as RSDformer. First, given the irregular shapes and nonuniform distributions of haze in RS images, capturing both local and non-local features is crucial for RS image dehazing models. Hence, we propose a detail-compensated transposed attention (DCTA) to extract the global and local dependencies (LDs) across channels. Second, to enhance the ability to learn degraded features and better guide the restoration process, we develop a dual-frequency adaptive block (DFAB) with dynamic filters. Finally, a dynamic gated fusion block (DGFB) is designed to achieve fuse and exchange features across different scales effectively. In this way, the model exhibits robust capabilities to capture dependencies from both global and local areas, resulting in improving image content recovery. Extensive experiments prove that the proposed method obtains more appealing performances against other competitive methods. [ABSTRACT FROM AUTHOR]

Images degraded by light scattering and absorption, such as hazy, sandstorm, and underwater images, often suffer color distortion and low contrast because of light traveling through turbid media. In order to enhance and restore such images, we first estimate ambient light using the depth-dependent color change. Then, via calculating the difference between the observed intensity and the ambient light, which we call the scene ambient light differential, scene transmission can be estimated. Additionally, adaptive color correction is incorporated into the image formation model (IFM) for removing color casts while restoring contrast. Experimental results on various degraded images demonstrate the new method outperforms other IFM-based methods subjectively and objectively. Our approach can be interpreted as a generalization of the common dark channel prior (DCP) approach to image restoration, and our method reduces to several DCP variants for different special cases of ambient lighting and turbid medium conditions.

In order to avoid the noise diffusion and amplification caused by traditional dehazing algorithms, a single image haze removal algorithm based on nonsubsampled contourlet transform (HRNSCT) is proposed. The HRNSCT removes haze only from the low-frequency components and suppresses noise in the high-frequency components of hazy images, preventing noise amplification caused by traditional dehazing algorithms. First, the nonsubsampled contourlet transform (NSCT) is used to decompose each channel of a hazy and noisy color image into low-frequency sub-band and high-frequency direction sub-bands. Second, according to the low-frequency sub-bands of the three channels, the color attenuation prior and dark channel prior are combined to estimate the transmission map, and use the transmission map to dehaze the low frequency sub-bands. Then, to achieve the noise suppression and details enhancement of the dehazed image, the high-frequency direction sub-bands of the three channels are shrunk, and those shrunk sub-bands are enhanced according to the transmission map. Finally, the nonsubsampled contourlet inverse transform is performed on the dehazed low-frequency sub-bands and enhanced high-frequency sub-bands to reconstruct the dehazed and noise-suppressed image. The experimental results show that the HRNSCT provides excellent haze removal and noise suppression performance and prevents noise amplification during dehazing, making it well suited for removing haze from noisy images.

In order to alleviate adverse impacts of haze on high-level vision tasks, image dehazing attracts great attention from computer vision research field in recent years. Most of existing methods are grouped into physical prior based and non-physical data-driven based categories. However, image dehazing is a challenging ill-conditioned and inherently ambiguous problem. Due to random distribution and concentration of haze, color distortion and excessive brightness often happen in physical prior based methods. Defects on high-frequency details' recovery are not solved well in non-physical data-driven methods. Therefore, to overcome these obstacles, in this paper, we have proposed an effective progressive back-traced dehazing network based on multi-resolution recurrent reconstruction strategies. A kind of irregular multi-scale convolution module is proposed to extract fine-grain local structures. And a kind of multi-resolution residual fusion module is proposed to progressively reconstruct intermediate haze-free images. We have compared our method with several popular state-of-the-art methods on public RESIDE and 2018 NTIRE Dehazing datasets. The experiment results demonstrate that our method could restore satisfactory high-frequency textures and high-fidelity colors. Related source code and parameters will be distributed on Github for further study.

Atmospheric scattering model (ASM) is one of the most widely used model to describe the imaging processing of hazy images. However, we found that ASM has an intrinsic limitation which leads to a dim effect in the recovered results. In this paper, by introducing a new parameter, i.e., light absorption coefficient, into ASM, an enhanced ASM (EASM) is attained, which can address the dim effect and better model outdoor hazy scenes. Relying on this EASM, a simple yet effective gray-world-assumption-based technique called IDE is then developed to enhance the visibility of hazy images. Experimental results show that IDE eliminates the dim effect and exhibits excellent dehazing performance. It is worth mentioning that IDE does not require any training process or extra information related to scene depth, which makes it very fast and robust. Moreover, the global stretch strategy used in IDE can effectively avoid some undesirable effects in recovery results, e.g., over-enhancement, over-saturation, and mist residue, etc. Comparison between the proposed IDE and other state-of-the-art techniques reveals the superiority of IDE in terms of both dehazing quality and efficiency over all the comparable techniques. [ABSTRACT FROM AUTHOR]

Images/videos captured from outdoor visual devices are usually degraded by turbid media, such as haze, smoke, fog, rain, and snow. Haze is the most common one in outdoor scenes due to the atmosphere conditions. In this paper, a novel deep learning-based architecture (denoted by MSRL-DehazeNet) for single image haze removal relying on multi-scale residual learning (MSRL) and image decomposition is proposed. Instead of learning an end-to-end mapping between each pair of hazy image and its corresponding haze-free one adopted by most existing learning-based approaches, we reformulate the problem as restoration of the image base component. Based on the decomposition of a hazy image into the base and the detail components, haze removal (or dehazing) can be achieved by both of our multi-scale deep residual learning and our simplified U-Net learning only for mapping between hazy and haze-free base components, while the detail component is further enhanced via the other learned convolutional neural network (CNN). Moreover, benefited by the basic building block of our deep residual CNN architecture and our simplified U-Net structure, the feature maps (produced by extracting structural and statistical features), and each previous layer can be fully preserved and fed into the next layer. Therefore, possible color distortion in the recovered image would be avoided. As a result, the final haze-removed (or dehazed) image is obtained by integrating the haze-removed base and the enhanced detail image components. Experimental results have demonstrated good effectiveness of the proposed framework, compared with state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

In this study, we introduce an algorithm which is based on a series of wellknown algorithms and mainly uses an improved dark channel prior algorithm and the White-Patch Retinex algorithm (both are heterogeneous algorithms) in order to effectively remove the haze from a single image. When used in conjunction with a heterogeneous architecture, the value of the algorithm becomes even greater. With an effective design and a novel procedure, the proposed algorithm can not only restore a clear image, but also solve the halo effect, color distortion, and long operating time issues resulting from the dark channel prior. Rich experimental results (visually and numerically) confirm that the performance and effectiveness of the new algorithm are better than those of the dark channel prior algorithm, indicating the applicability of the new algorithm. [ABSTRACT FROM AUTHOR]

Haze often limits visibility and reduces contrast in outdoor images. The degradation varies spatially since it depends on the objects’ distances from the camera. This dependency is expressed in the transmission coefficients, which control the attenuation. Restoring the scene radiance from a single image is a highly ill-posed problem, and thus requires using an image prior. Contrary to methods that use patch-based image priors, we propose an algorithm based on a non-local prior. The algorithm relies on the assumption that colors of a haze-free image are well approximated by a few hundred distinct colors, which form tight clusters in RGB space. Our key observation is that pixels in a given cluster are often non-local, i.e., spread over the entire image plane and located at different distances from the camera. In the presence of haze these varying distances translate to different transmission coefficients. Therefore, each color cluster in the clear image becomes a line in RGB space, that we term a haze-line. Using these haze-lines, our algorithm recovers the atmospheric light, the distance map and the haze-free image. The algorithm has linear complexity, requires no training, and performs well on a wide variety of images compared to other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

For undersea activities, underwater vehicles usually carry optical imaging systems for recording. The captured images and videos frequently suffer from two displeasing problems: First, color distortion; and second, poor visibility. This is mainly because that the light is exponentially attenuated while penetrating through water and the strength of attenuation is color dependent. This paper develops an effective single underwater image restoration framework based on the depth estimation and the transmission compensation. To address the consequences of scattering and absorption, our restoration scheme consists of five major phases: 1) background light estimation; 2) submerged dark channel prior; 3) transmission refinement and radiance recovery; 4) point spread function deconvolution; and 5) transmission and color compensation. A wide variety of underwater images with various scenarios were exploited to assess the restoration performance of the proposed algorithm. Various evaluation metrics were employed to analyze the experimental results. It was suggested that this new restoration algorithm outperformed many state-of-the-art methods both qualitatively and quantitatively. In addition, potential applications regarding autopilot and three-dimensional visualization were demonstrated. We believe that our underwater image restoration technique is promising in many undersea activities that require high-quality images. [ABSTRACT FROM AUTHOR]

Haze removal typically works on a physical model to estimate how light is transmitted and lost due to absorption and scattering through the atmosphere. In this paper, a region detection network is proposed to learn the relationship between the hazy image and the medium transmission map in a patchwise manner; the transmission map is then used to remove haze via an atmospheric scattering model and enhance the detail of de-hazed images. To this end, we design a simple yet powerful deep convolutional neural network, which mainly consists of two types of network units and can be trained in an end-to-end manner. One network unit is a module with the residual structure that facilitates the learning process of the deep network. The other is a novel module with a cascaded cross channel pool, which fuses multi-level haze-relevant features and boosts the abstraction ability of the model on a nonlinear manifold. Moreover, an evolutionary-based enhancement method is developed to improve the level of detail of over-smoothed results. Several comparative experiments have been conducted on synthetic and real images, through which we conclude that the proposed method achieves state-of-the-art haze removal results, qualitatively and quantitatively. Supplementary experiments further indicate that our method works better against other adverse effects on vision quality (e.g., the mist formed by heavy rain and the veil met underwater). Moreover, we present a lightweight version of the proposed network, which achieves an impressive haze removal performance even on low-power devices. [ABSTRACT FROM AUTHOR]

Abstract Restoring hazy images is challenging since it must account for several physical factors that are related to the image formation process. Existing analytical methods can only provide partial solutions because they rely on assumptions that may not be valid in practice. This research presents an effective method for restoring hazy images based on genetic programming. Using basic mathematical operators several computer programs that estimate the medium transmission function of hazy scenes are automatically evolved. Afterwards, image restoration is performed using the estimated transmission function in a physics-based restoration model. The proposed estimators are optimized with respect to the mean-absolute-error. Thus, the effects of haze are effectively removed while minimizing overprocessing artifacts. The performance of the evolved GP estimators given in terms of objective metrics and a subjective visual criterion, is evaluated on synthetic and real-life hazy images. Comparisons are carried out with state-of-the-art methods, showing that the evolved estimators can outperform these methods without incurring a loss in efficiency, and in most scenarios achieving improved performance that is statistically significant. [ABSTRACT FROM AUTHOR]

Outdoor images captured during hazy conditions have degraded visibility. The lack of both a medium transmission and atmospheric lights in a single haze image cause an ill-posed problem in the atmospheric scattering model. This paper proposes a novel haze density estimation model with a universal atmospheric-light extractor for single-image dehazing. The proposed method employs exponential inference to construct an exponential inference model to more accurately estimate haze density compared with the state-of-the-art methods. The coefficients in the proposed haze density estimation model are learned using a turbulent particle swarm optimization technique to obtain the best approximation of medium transmission. Moreover, a novel universal atmospheric-light extractor based on support vector data description is utilized to resolve the problem caused by a lack of atmospheric light. The overall results obtained by conducting qualitative and quantitative evaluations demonstrated that the proposed method has substantially higher dehazing efficacy and produces fewer artifacts than the state-of-the-art haze removal methods. [ABSTRACT FROM AUTHOR]

A framework for image visibility restoration and haze removal is proposed. The proposed technique utilizes hybrid median filtering in conjunction with accelerated local Laplacian filtering for initial dehazing of images. For visual enhancement and correct restoration of colors, constrained l0 -based gradient image decomposition is applied. The proposed technique not only effectively removes haze from the images but also addresses the issues of distorted colors, visual, and halo artifacts, and haze removal from sky region in images in a better way when compared to other techniques. Experiments were performed on outdoor RGB images as well as remotely sensed images. The effectiveness of our proposed technique is demonstrated by quantitative and visual analyzes. [ABSTRACT FROM PUBLISHER]

Multispectral remote sensing images are often contaminated by haze, which causes low image quality. In this paper, a novel dehazing method based on a deep convolutional neural network (CNN) with the residual structure is proposed for multispectral remote sensing images. First, multiple CNN individuals with the residual structure are connected in parallel and each individual is used to learn a regression from the hazy image to the clear image. Then, the outputs of CNN individuals are fused with weight maps to produce the final dehazing result. In the designed network, the CNN individuals, mining multiscale haze features through multiscale convolutions, are trained using different levels of haze samples to achieve different dehazing abilities. In addition, the weight maps change with the haze distribution, and the fusion of the CNN individuals is adaptive. The designed network is end-to-end, and putting a hazy image into it, the clear scene can be restored. To train the network, a wavelength-dependent haze simulation method is proposed to generate labeled data, which can synthesize hazy multispectral images highly close to real conditions. Experimental results show that the proposed method can accurately remove the haze in each band of multispectral images under different scenes. [ABSTRACT FROM PUBLISHER]

This paper presents a convolutional neural network to automatically remove the haze distribution using a single multispectral remote sensing image in the raw file format. To train the proposed dehazing network, we synthesized multispectral hazy images using the haze thickness map (HTM) and relative scattering model representing the wavelength-dependent scattering property of the haze distribution. Since the raw multispectral hazy images have a low dynamic range, we cannot accurately estimate the haze distribution directly from them. To differently impose a proper amount of attention to hazy and haze-free regions, we used the HTM from the contrast-enhanced version of the input hazy image. The proposed dehazing network consists of four sub-networks: (i) shallow feature extraction network (SFEN), (ii) cascaded residual dense block network (CRDBN), (iii) multiscale feature extraction network (MFEN), and (iv) refinement network (RN). The densely connected convolutional layers and local residual learning allow the residual dense block (RDB) to extract the abundant local features, and the cascaded architecture further improves the propagation of the local information and gradients. The MFEN is used to extract multiscale local features representing the hierarchical information for the haze distribution and haze-free region. Experimental results demonstrated that the proposed method can achieve improved dehazing performance on Korea Multi-Purpose Satellite-3A (KOMPSAT-3A) multispectral remote sensing imagery without undesired artifacts. In the sense of quantitative assessment, the proposed method produced improved peak signal-to-noise ratio (PSNR) by 10%, structural similarity index measure (SSIM) by 1%, and spectral angle mapper (SAM) by 19% compared with the existing best method. [Display omitted] • Novel method generates multispectral hazy images using scattering coefficient and haze thickness map. • Proposed network leverages local and global context with cascaded residual dense block. • HTM of contrast-enhanced images provides selective attention to hazy and haze-free regions. • Experiment with KOMPSAT-3A images reduces over-dehazing and spectral distortion artifacts. • Proposed method improves PSNR by 10%, SSIM by 1%, and SAM by 19% compared to existing methods. [ABSTRACT FROM AUTHOR]

Image haze removal has become an important research direction in the field of computer vision because of the vast development and increasing demand of its applications. In the present era, where uploading images in social media has become very common, the image editing and processing applications has gained much popularity. Outdoor images that are captured in bad weather are seriously degraded due to several factors. Haze, being the most common degrading factor, affects the visibility of the images and makes them unclear. Haze is formed due to a combination of two fundamental phenomena namely, attenuation and airlight. Attenuation deteriorates the scene contrast and air light increases the whiteness in the scene. Therefore, the removal of attenuation and airlight helps to restore the color and contrast of the images, making them haze-free. Different works have been proposed till date inorder to filter out haze from images. This paper analyses the different image haze removal techniques. The different techniques, along with their merits and demerits are discussed. [ABSTRACT FROM AUTHOR]

Haze removal is an important image restoration technology that aims to remove annoying haze particles from images. However, the efficacies of traditional dehazing methods are easily hindered by insufficient estimation of haze thickness, and thus cannot effectively provide satisfactory haze removal results. In this paper, we propose an edge collapse-based dehazing algorithm by which to dynamically repair the transmission map and, thereby, achieve satisfactory visibility restoration. Experimental results using qualitative and quantitative evaluations demonstrate that the haze removal ability of the proposed edge collapse-based dehazing method is significantly superior to those of other state-of-the-art methods. [ABSTRACT FROM PUBLISHER]

Haze can turn a colored image into a white-and-gray one, causing a loss of image detail and a reduction in contrast. Haze also disturbs many applications, including direct target observations and indirect recognition, tracking, and measurements. Image dehazing can remove haze from images, increase the scene visibility, and improve the overall visual effect. However, most existing dehazing methods, which require a large number of computations and complex steps, cannot meet the requirements of a real-time application. In this paper, a fast single-image dehazing algorithm is proposed. Using the proposed haze imaging model, we estimate the atmospheric light using an infinite sky region and close white region, calculate the transmission map using a linear transformation, and recover a haze-free image. We also demonstrate its good performance by conducting experiments on various hazy images. The proposed algorithm can be efficiently implemented in O( N ) time and has a faster speed than most previous dehazing methods. In addition, the method takes no more than 40 ms to process an image of one-million pixels on a PC with a 3.1 GHz Intel Xeon processor. [ABSTRACT FROM AUTHOR]

In order to improve contrast and visibility of haze image captured in the foggy weather, this paper proposed an image restoration algorithm using domain transform recursive filter. Based on the atmospheric scattering model, the proposed method firstly reduced the image dimensions of physical constraint through domain transform, and then obtained the accurate atmospheric veil by performing the recursive filter. Finally the clear image was recovered by solving the physical equation of haze image and enhanced by a local linear mapping. The proposed algorithm not only got more natural effectiveness at the edge of places where scene depth changed abruptly, but also effectively highlighted detail information of recovered image. Experimental results show that comparing with traditional single image haze removal methods, the proposed method has better visual effect and faster execution speed. Furthermore, the algorithm can be performed in parallel, so it can be accelerated using GPU and satisfy real-time applications. [ABSTRACT FROM AUTHOR]