Dehazing Methods Research Articles

Person re-identification using adversarial haze attack and defense: A deep learning framework

In this paper, the adversarial haze attack problem is addressed using the dark channel prior (DCP) de-hazing method. The adversarial attack affects rank-1 accuracy, where searching a target image against each test image is a specific search problem. To resolve this kind of problem, a feature fusion model is proposed to fuse handcrafted features and a pre-trained network model to obtain robust and discriminative features. The proposed model learns global features using transfer learning architecture whereas local features are obtained using the conventional method. Three pre-trained CNN models (AlexNet, ResNet, and Inception-v3) are used for feature extraction via transfer learning. The experiments are performed on publicly available datasets, achieving 68.6% accuracy in rank-1 with VIPER dataset and 79.6% accuracy with CHUK03 dataset. The proposed model enhances rank-1 accuracy of person re-identification when comparing with other state-of-the-art methods.

Single Image Dehazing Using Sparse Contextual Representation

In this paper, we propose a novel method to remove haze from a single hazy input image based on the sparse representation. In our method, the sparse representation is proposed to be used as a contextual regularization tool, which can reduce the block artifacts and halos produced by only using dark channel prior without soft matting as the transmission is not always constant in a local patch. A novel way to use dictionary is proposed to smooth an image and generate the sharp dehazed result. Experimental results demonstrate that our proposed method performs favorably against the state-of-the-art dehazing methods and produces high-quality dehazed and vivid color results.

Photo-Realistic Image Dehazing and Verifying Networks via Complementary Adversarial Learning.

Physical model-based dehazing methods cannot, in general, avoid environmental variables and undesired artifacts such as non-collected illuminance, halo and saturation since it is difficult to accurately estimate the amount of the illuminance, light transmission and airlight. Furthermore, the haze model estimation process requires very high computational complexity. To solve this problem by directly estimating the radiance of the haze images, we present a novel dehazing and verifying network (DVNet). In the dehazing procedure, we enhanced the clean images by using a correction network (CNet), which uses the ground truth to learn the haze network. Haze images are then restored through a haze network (HNet). Furthermore, a verifying method verifies the error of both CNet and HNet using a self-supervised learning method. Finally, the proposed complementary adversarial learning method can produce results more naturally. Note that the proposed discriminator and generators (HNet & CNet) can be learned via an unpaired dataset. Overall, the proposed DVNet can generate a better dehazed result than state-of-the-art approaches under various hazy conditions. Experimental results show that the DVNet outperforms state-of-the-art dehazing methods in most cases.

Attention-based Feature Enhanced Dehazing Network

Presence of haze in images obscures underlying information, which is undesirable in applications requiring accurate environment information. To recover such an image, a dehazing algorithm should enhance the feature information of the background while weakening the feature information of haze. In this paper, we propose an end-to-end attention-based feature enhanced dehazing network (AEDNet), which integrates enhancement strategy and attention mechanism, to achieve haze removal. The network is based on U-Net, which has the advantages of retaining information, obtaining multi-scale features and so on. In the training of the network, pixel loss and perceptual loss are used to preserve feature information and improve the overall quality of results. The extensive evaluation shows that the proposed model performs significantly better than previous dehazing methods on various benchmarks.

GGADN: Guided generative adversarial dehazing network

Image dehazing has always been a challenging topic in image processing. The development of deep learning methods, especially the generative adversarial networks (GAN), provides a new way for image dehazing. In recent years, many deep learning methods based on GAN have been applied to image dehazing. However, GAN has two problems in image dehazing. Firstly, For haze image, haze not only reduces the quality of the image but also blurs the details of the image. For GAN network, it is difficult for the generator to restore the details of the whole image while removing the haze. Secondly, GAN model is defined as a minimax problem, which weakens the loss function. It is difficult to distinguish whether GAN is making progress in the training process. Therefore, we propose a guided generative adversarial dehazing network (GGADN). Different from other generation adversarial networks, GGADN adds a guided module on the generator. The guided module verifies the network of each layer of the generator. At the same time, the details of the map generated by each layer are strengthened. Network training is based on the pre-trained VGG feature model and L1-regularized gradient prior which is developed by new loss function parameters. From the dehazing results of synthetic images and real images, the proposed method is better than the state-of-the-art dehazing methods.

Single image mixed dehazing method based on numerical iterative model and DehazeNet.

As one of the most common adverse weather phenomena, haze has caused detrimental effects on many computer vision systems. To eliminate the effect of haze, in the field of image processing, image dehazing has been studied intensively, and many advanced dehazing algorithms have been proposed. Physical model-based and deep learning-based methods are two competitive methods for single image dehazing, but it is still a challenging problem to achieve fidelity and effectively dehazing simultaneously in real hazy scenes. In this work, a mixed iterative model is proposed, which combines a physical model-based method with a learning-based method to restore high-quality clear images, and it has good performance in maintaining natural attributes and completely removing haze. Unlike previous studies, we first divide the image into different regions according to the density of haze to accurately calculate the atmospheric light for restoring haze-free images. Then, dark channel prior and DehazeNet are used to jointly estimate the transmission to promote the final clear haze-free image that is more similar to the real scene. Finally, a numerical iterative strategy is employed to further optimize the atmospheric light and transmission. Extensive experiments demonstrate that our method outperforms existing state-of-the-art methods on synthetic datasets and real-world datasets. Moreover, to indicate the universality of the proposed method, we further apply it to the remote sensing datasets, which can also produce visually satisfactory results.

RGNAM: recurrent grid network with an attention mechanism for single-image dehazing

Single-image dehazing is a critical problem since haze existence degrades the quality of images and hinders most advanced computer vision tasks. Early methods solve this problem via the atmospheric scattering model, which estimate the intermediate parameters and then recover a clear image by low-level priors or learning on synthetic datasets. However, these model-based methods do not hold in various scenes. Recently, many learning-based methods have directly recovered dehazed images from inputs, but these methods fail to deal with dense haze and always lead to color distortion. To solve this problem, we build a recurrent grid network with an attention mechanism, named RGNAM. Specifically, we propose a recurrent feature extraction block, which repeats a local residual structure to enhance feature representation and adopts a spatial attention module to focus on dense haze. To alleviate color distortion, we extract local features (e.g., structures and edges) and global features (e.g., colors and textures) from a grid network and propose a feature fusion module combining trainable weights and channel attention mechanisms to merge these complementary features effectively. We train our model with smooth L1 loss and structural similarity loss. The experimental results demonstrate that our proposed RGNAM surpasses previous state-of-the-art single-image dehazing methods on both synthetic and real haze datasets.

New approach to dehaze single nighttime image

This paper focus on the dehazing of a single image captured at nighttime. The current state-of-the-art nighttime dehazing approaches usually suffer from the color shift problem due to the fact that the assumptions enforced underdaytime cannot get applied to the nighttime image directly. The classical dehazing methods try to estimate the transmission mapand accurate light to dehaze a single image. The present basic idea is to firstly separate the light layer from the hazy image and thetransmission map can be computed afterwards. A new layer separation method is proposed to solve the non-globalatmospheric light problem. The present method on some real datasets to show its superior performance is validated.

Single image dehazing based on haze density estimation in different color spaces

In this paper, an efficient method for single image dehazing is proposed based on haze density estimation. We provide two forms of haze density estimation in different color spaces, which are called scene-based haze density estimation in HSV color space and pixel-based haze density estimation in RGB color space. The attenuation model of pixel-level transmission is established based on the two haze density estimations by an exponential function. Guided filtering is applied to smooth the transmission map and maintain the local edges. Global atmospheric light is obtained adaptively by smoothed transmission. A series of experiments on different types of hazy images are implemented, and the results reveal that the proposed method can obtain high-quality haze-free images along with abundant details, high color fidelity, and few halo artifacts.

Deep Learning-Enabled Variational Optimization Method for Image Dehazing in Maritime Intelligent Transportation Systems

Image dehazing has become a fundamental problem of common concern in computer vision-driven maritime intelligent transportation systems (ITS). The purpose of image dehazing is to reconstruct the latent haze-free image from its observed hazy version. It is well known that the accurate estimation of transmission map plays a vital role in image dehazing. In this work, the coarse transmission map is firstly estimated using a robust fusion-based strategy. A unified optimization framework is then proposed to estimate the refined transmission map and latent sharp image simultaneously. The resulting constrained minimization model is solved using a two-step optimization algorithm. To further enhance dehazing performance, the solutions of subproblems obtained in this optimization algorithm are equivalent to deep learning-based image denoising. Due to the powerful representation ability, the proposed method can accurately and robustly estimate the transmission map and latent sharp image. Numerous experiments on both synthetic and realistic datasets have been performed to compare our method with several state-of-the-art dehazing methods. Dehazing results have demonstrated the proposed method’s superior imaging performance in terms of both quantitative and qualitative evaluations. The enhanced imaging quality is beneficial for practical applications in maritime ITS, for example, vessel detection, recognition, and tracking.

A joint cumulative distribution function and gradient fusion based method for dehazing of long shot hazy images

Hazy or foggy weather conditions significantly degrade the visual quality of an image in an outdoor environment. It also changes the color and reduces the contrast of an image. This paper introduces a novel single image dehazing technique to restore a hazy image without considering the physical model of haze formation. In order to find haze-free image, the proposed method does not require the transmission map and its costly refinement process. Since haze effect is dependent on the depth, it severely degrades the visibility of the objects located at a far distance. The objects close to the camera are unaffected. In this paper, we propose a fusion-based haze removal method based on the joint cumulative distribution function (JCDF) that treats faraway haze and nearby haze separately. The output images after the JCDF module, fused in the gradient domain to produce a haze-free image. The proposed method not only significantly enhances visibility but also preserves texture details. The proposed method is experimented and evaluated on a large set of challenging hazy images (large scene depth, night time, dense fog, etc.). Both qualitative and quantitative measures show that the performance of the proposed method is better than the state-of-the-art dehazing techniques.

Review of passive polarimetric dehazing methods

The images acquired in haze conditions are significantly degraded due to the presence of atmospheric particles. These images have low contrast, poor visibility, and some information is lost as well. These characteristics severely hinder the further processing and the applications in which the images are being used. The polarimetric dehazing methods are effective for direct imaging and enhancing the imaging quality. In addition, polarimetric dehazing methods have the capability to cope with haze and other turbid mediums. Therefore, the polarimetric dehazing methods are extensively developed and used in different applications due to their superior performance. We present in detail the principles, the implementation techniques, and the advancements in the polarimetric dehazing methods. We believe that this work is the first in-depth review on the passive polarimetric dehazing methods.

A Robust Method for Dehazing of Single Image with Sky Region Detection and Segmentation

In recent times, there has been a tremendous progress in image dehazing for computer vision applications, while the sky region processed by these algorithms tends to degrade by noise and color distortion. In this paper, an improved dark channel prior algorithm is proposed which detects the sky region first and divides the image into sky region and non-sky region and then estimates the transmission of two parts separately, followed by combining with refining step. The proposed algorithm also accurately corrects the transmission of the sky region to avoid noise and color distortion. Experimental results show a greater quality improvement in the output images than the existing strategies.

Range-intensity-profile prior dehazing method for underwater range-gated imaging

This paper is concerned with the mitigation of backscatter effects in a single gated image. A range-intensity-profile prior dehazing method is proposed to estimate scene depth and finely remove water backscatter at different depths for underwater range-gated imaging. It is based on the prior that the target intensity is distributed with range intensity profiles in gated images. The depth transmission and depth-noise map are then calculated from the scene depth. A high-quality image is restored by subtracting the depth-noise map and dividing the depth transmission. The simulation and experimental results show that the proposed method works well even if a portion of the estimated depth may be smaller than its real value, and the peak signal-to-noise ratio of dehazing images gets up to a doubled increase.

Single Color Image De-Hazing Using Dark Channel Prior

In real world scenario due to bad weather conditions the presence of fog and haze, the particles in the outdoor environment or atmosphere (e.g., droplets, smoke, sand, snow, mist, volcanic ash, liquid dust or solid dust) greatly reduces the visibility of the scene. As a consequence, the clarity of an image would be seriously degraded, which may decrease the performance of many image processing applications. Image Dehazing methods try to alleviate these problems by estimating a haze free version of the given hazy image. Traditionally the task of image dehazing can be processed as recovering the scene radiance from a noisy hazy image by estimating the atmospheric light and transmission properties. In those kinds of techniques, it additionally needs some more information regarding the image such as scene depth, weather condition parameters and so on. But this is not suitable for real world scenario. This research focus on proposing an approach to fully capture the intrinsic attributes of a hazy image and improves the performance of dehazing. Dark Channel Prior plays vital role in dehazing process. Hence this research focus on recovering dehaze version of the input image by CNN. So that all methods are comes under the categories image enhancement, image fusion image restoration based on statistical and structural features of the hazed image.

Research on Remote Sensing Image De‐haze Based on GAN

Commonly used remote sensing image de-haze methods include: the image enhancement method and a physical model-based. However, when the above methods are applied to high-resolution remote sensing images, problems with texture information loss and insufficient enhancement often occur. These problems affect further analysis and application of high-resolution remote sensing images. This paper proposes a new single-image de-haze method called texture attention GAN. In this network, in order to solve the problem of texture information loss in the process of de-haze, a texture attention-based generator is adopted. When design the network discriminator, the global and local discriminators are used to improve the distortion of image details. In comparison with several common methods, this method has achieved better results.

Region-Based Dehazing via Dual-Supervised Triple-Convolutional Network

Most physical model-based dehazing methods are subject to contrast degradation in a dark or shadow region because of the mismatch between the physical model and real haze. This degradation decreases the quality of dehazed images. Furthermore, the retinex-haze combined models can cause the brightness saturation problem in a haze region. For this reason, the retinex-haze combined approaches are not appropriate to enhance the real-world haze images. To solve these problems, we present a novel region-based dehazing method via dual-supervised triple-convolutional network (TCN). More specifically, the proposed network first simulates the mismatch problem based on the region-model. Next, we then train the proposed triple-convolutional network, which can estimate the degraded regions. We then present a novel dual-supervised learning method to efficiently train the networks using a non-ideal dataset. Experimental results show that the proposed method outperforms state-of-the-art approaches in solving complex haze problems. The output of the proposed network has a high-similarity index in most cases for various benchmark dataset. Our approach also produces high-quality images in real haze image datasets.

Distortion-free image dehazing by superpixels and ensemble neural network

Single image dehazing is a technique used to remove the effect of haze from an image captured in poor weather conditions. Due to the scattering of particles, a captured image suffers from low visibility and contrast. Besides, scattering also adds nonlinear noise to the captured image. Existing image dehazing methods improve the visibility of the hazy image. However, these methods significantly generate artifacts such as halo at the depth discontinuities, blocking, and color aliasing in the sky regions. Some methods addressed this problem, but these methods introduce other issues such as loss of details, blurring effects, and oversaturation in the dehazed image. This paper proposes a method using superpixels and ensemble nonlinear regression to estimate the transmission that improves the visibility of a hazy image without any artifact. Conventional machine learning methods require a vast amount of haze-free and hazy images of different haze concentrations to train the model. The use of superpixels offers less number of training examples and also helps in reducing halo artifacts. The ensemble nonlinear regression predicts the transmission for a superpixel in such a way that the recovered image looks more natural, especially in the sky regions. The proposed method is evaluated by the various distortion parameters on real-world challenging and synthetic hazy images. The qualitative and quantitative analysis in experimental results proves that the proposed method is superior to that of state-of-the-art dehazing methods.

Indirect Domain Shift for Single Image Dehazing

Despite their remarkable expressibility, convolution neural networks (CNNs) still fall short of delivering satisfactory results on single image dehazing, especially in terms of faithful recovery of fine texture details. In this paper, we argue that the inadequacy of conventional CNN-based dehazing methods can be attributed to the fact that the domain of hazy images is too far away from that of clear images, rendering it difficult to train a CNN for learning direct domain shift through an end-to-end manner and recovering texture details simultaneously. To address this issue, we propose to add explicit constraints inside a deep CNN model to guide the restoration process. In contrast to direct learning, the proposed mechanism shifts and narrows the candidate region for the estimation output via multiple confident neighborhoods. Therefore, it is capable of consolidating the expressibility of different architectures, resulting in a more accurate indirect domain shift (IDS) from the hazy images to that of clear images. We also propose two different training schemes, including hard IDS and soft IDS, which further reveal the effectiveness of the proposed method. Our extensive experimental results indicate that the dehazing method based on this mechanism dramatically outperforms the state-of-the-arts.

Recurrent Context Aggregation Network for Single Image Dehazing

Existing learning-based dehazing methods are prone to cause excessive dehazing and failure to dense haze, mainly because that the global features of hazy images are not fully utilized, while the local features of hazy images are not enough discriminative. In this letter, we propose a Recurrent Context Aggregation Network (RCAN) to effectively dehaze images and restore color fidelity. In RCAN, an efficient and generic module, called Context Aggression Block (CAB), is designed to improve the feature representation by taking advantage of both global and local features, which are complementary for robust dehazing because that local features can capture different levels of haze, and global features can focus on textures and object edges of a whole image. In addition, RCAN adopts a deep recurrent mechanism to improve the dehazing performance without introducing additional network parameters. Extensive experimental results on both synthetic and real-world datasets show that the proposed RCAN performs better than other state-of-the-art dehazing methods.