Single Image Defogging Research Articles

Single Image Atmospheric Veil Removal Using New Priors for Better Genericity

From an analysis of the priors used in state-of-the-art algorithms for single image defogging, a new prior is proposed to obtain a better atmospheric veil removal. Our hypothesis is based on a physical model, considering that the fog appears denser near the horizon rather than close to the camera. It leads to more restoration when the fog depth is more important, for a more natural rendering. For this purpose, the Naka–Rushton function is used to modulate the atmospheric veil according to empirical observations on synthetic foggy images. The parameters of this function are set from features of the input image. This method also prevents over-restoration and thus preserves the sky from artifacts and noises. The algorithm generalizes to different kinds of fog, airborne particles, and illumination conditions. The proposed method is extended to the nighttime and underwater images by computing the atmospheric veil on each color channel. Qualitative and quantitative evaluations show the benefit of the proposed algorithm. The quantitative evaluation shows the efficiency of the algorithm on four databases with different types of fog, which demonstrates the broad generalization allowed by the proposed algorithm, in contrast with most of the currently available deep learning techniques.

Defogging Technology Based on Dual-Channel Sensor Information Fusion of Near-Infrared and Visible Light

Since the method to remove fog from images is complicated and detail loss and color distortion could occur to the defogged images, a defogging method based on near-infrared and visible image fusion is put forward in this paper. The algorithm in this paper uses the near-infrared image with rich details as a new data source and adopts the image fusion method to obtain a defog image with rich details and high color recovery. First, the colorful visible image is converted into HSI color space to obtain an intensity channel image, color channel image, and saturation channel image. The intensity channel image is fused with a near-infrared image and defogged, and then it is decomposed by Nonsubsampled Shearlet Transform. The obtained high-frequency coefficient is filtered by preserving the edge with a double exponential edge smoothing filter, while low-frequency antisharpening masking treatment is conducted on the low-frequency coefficient. The new intensity channel image could be obtained based on the fusion rule and by reciprocal transformation. Then, in color treatment of the visible image, the degradation model of the saturation image is established, which estimates the parameters based on the principle of dark primary color to obtain the estimated saturation image. Finally, the new intensity channel image, the estimated saturation image, and the primary color image are reflected to RGB space to obtain the fusion image, which is enhanced by color and sharpness correction. In order to prove the effectiveness of the algorithm, the dense fog image and the thin fog image are compared with the popular single image defogging and multiple image defogging algorithms and the visible light-near infrared fusion defogging algorithm based on deep learning. The experimental results show that the proposed algorithm is better in improving the edge contrast and the visual sharpness of the image than the existing high-efficiency defogging method.

Improved color attenuation prior based image de-fogging technique

Single image de-fogging has been a confronting problem due to its ill-posed nature. In this paper, a de-fogging technique based upon color attenuation prior (CAP) has been proposed. CAP uses linear model and learning the parameters of this model with a supervised learning method for estimating the scene depth of a foggy image. This depth map is utilized for transmission map estimation. Transmission (t) is one of the important parameter of physical model based de-fogging techniques which describes the portion of the light coming from the scene point that is not scattered and reaches the camera. It is a map called transmission or transparency of the fog 0 < t(x) < 1, t(x) = 0 means completely foggy, t(x) = 1 means fog-free. The more accurately the transmission or depth is estimated, the better the defogging performance will be. In the proposed work, to quickly and accurately estimate the transmission map, a sub-sampling based local minimum operation and fast gradient domain guided image filtering (GDGF) is applied on CAP based initial depth map. The edge attentive restraints of GDGF make edges to be conserved better in the de-fogged images. The de-fogged images obtained by CAP technique suffer from dullness and higher illumination variations due to consideration of fog image degradation model in homogeneous environment and a constant value of atmospheric light. Such variations are removed in the proposed work by using Lambert’s law of illumination reflection, which helps to compensate non uniform illumination, causes simultaneous dynamic range modification, color consistency, and lightness rendition without producing the artifacts in a de-fogged image. To improve the processing speed, image sub-sampling mechanism is used in various steps of image de-fogging. The sub-sampling is used in such a way that the quality of the output at any step is not compromised as demonstrated through various quality parameters. Experimental results show that the proposed approach outperforms state-of-the-art fog removal techniques in terms of efficiency and the de-fogging effect.

End-to-End Single Image Fog Removal Using Enhanced Cycle Consistent Adversarial Networks

Single image defogging is a classical and challenging problem in computer vision. Existing methods towards this problem mainly include handcrafted priors based methods that rely on the use of the atmospheric degradation model and learning based approaches that require paired fog-fogfree training example images. In practice, however, prior-based methods are prone to failure due to their own limitations and paired training data are extremely difficult to acquire. Inspired by the principle of CycleGAN network, we have developed an end-to-end learning system that uses unpaired fog and fogfree training images, adversarial discriminators and cycle consistency losses to automatically construct a fog removal system. Similar to CycleGAN, our system has two transformation paths; one maps fog images to a fogfree image domain and the other maps fogfree images to a fog image domain. Instead of one stage mapping, our system uses a two stage mapping strategy in each transformation path to enhance the effectiveness of fog removal. Furthermore, we make explicit use of prior knowledge in the networks by embedding the atmospheric degradation principle and a sky prior for mapping fogfree images to the fog images domain. In addition, we also contribute the first real world nature fog-fogfree image dataset for defogging research. Our multiple real fog images dataset (MRFID) contains images of 200 natural outdoor scenes. For each scene, there are one clear image and corresponding four foggy images of different fog densities manually selected from a sequence of images taken by a fixed camera over the course of one year. Qualitative and quantitative comparison against several state-of-the-art methods on both synthetic and real world images demonstrate that our approach is effective and performs favorably for recovering a clear image from a foggy image.

A physics based generative adversarial network for single image defogging

In the field of single image defogging, there are two main methods. One is the image restoration method based on the atmospheric scattering theory which can recover the image texture details well. The other is the image enhancement method based on Retinex theory which can improve the image contrast well. In practice, however, the former can easily lead to low contrast images; the latter is prone to losing texture details. Therefore, how to effectively combine the advantages of both to remove fog is a key issue in the field. In this paper, we have developed a physics based generative adversarial network (PBGAN) to exploit the advantages between those two methods in parallel. To our knowledge, it is the first learning defogging framework that incorporates these two methods and to enable them to work together and complement each other. Our method has two generative adversarial modules, the Contrast Enhancement (CE) module and the Texture Restoration (TR) module. To improve contrast in the CE module, we introduced a novel inversion-adversarial loss and a novel inversion-cycle consistency loss for training the generator. To improve the texture in the TR module, we introduced two convolutional neural networks to learn the atmospheric light coefficient and the transmission map, respectively. Extensive experiments on both synthetic and real-world datasets demonstrate that the proposed approach performs better than several state-of-the-art methods quantitatively and qualitatively.

Single image defogging by gain gradient image filter

Single image defogging by gain gradient image filter

Fast Defogging and Restoration Assessment Approach to Road Scene Images

Single image defogging has received increased attention recently. However, most existing methods are generic restoration methods which may cause partial over-enhancing to the image at a short distance. This certainly makes these defogging methods may be not suitable for road images. In this paper, we propose a new image defogging algorithm that divides the original foggy image into three regions and different enhancement processes are performed for the different region. As the results, the far-away region image that is most concerned by a car driver can be moderately enhanced, and over-enhancement of the bottom region image can be also limited effectively. To effectively measure the defogging effects of various algorithms, the assessment methods based on image feature and visibility distance are proposed for road scene image. The comparative study and quantitative evaluation show that the proposed method can obtain very good defogging image with relatively fast speed.

Variational contrast enhancement guided by global and local contrast measurements for single-image defogging

The visibility of images captured in foggy conditions is impaired severely by a decrease in the contrasts of objects and veiling with a characteristic gray hue, which may limit the performance of visual applications out of doors. Contrast enhancement together with color restoration is a challenging mission for conventional fog-removal methods, as the degrading effect of fog is largely dependent on scene depth information. Nowadays, people change their minds by establishing a variational framework for contrast enhancement based on a physically based analytical model, unexpectedly resulting in color distortion, dark-patch distortion, or fuzzy features of local regions. Unlike previous work, our method treats an atmospheric veil as a scattering disturbance and formulates a foggy image as an energy functional minimization to estimate direct attenuation, originating from the work of image denoising. In addition to a global contrast measurement based on a total variation norm, an additional local measurement is designed in that optimal problem for the purpose of digging out more local details as well as suppressing dark-patch distortion. Moreover, we estimate the airlight precisely by maximization with a geometric constraint and a natural image prior in order to protect the faithfulness of the scene color. With the estimated direct attenuation and airlight, the fog-free image can be restored. Finally, our method is tested on several benchmark and realistic images evaluated by two assessment approaches. The experimental results imply that our proposed method works well compared with the state-of-the-art defogging methods.

Combining Stereo and Atmospheric Veil Depth Cues for 3D Reconstruction

Stereo reconstruction serves many outdoor applications, and thus sometimes faces foggy weather. The quality of the reconstruction by state of the art algorithms is then degraded as contrast is reduced with the distance because of scattering. However, as shown by defogging algorithms from a single image, fog provides an extra depth cue in the gray level of far away objects. Our idea is thus to take advantages of both stereo and atmospheric veil depth cues to achieve better stereo reconstructions in foggy weather. To our knowledge, this subject has never been investigated earlier by the computer vision community. We thus propose a Markov Random Field model of the stereo reconstruction and defogging problem which can be optimized iteratively using the α-expansion algorithm. Outputs are a dense disparity map and an image where contrast is restored. The proposed model is evaluated on synthetic images. This evaluation shows that the proposed method achieves very good results on both stereo reconstruction and defogging compared to standard stereo reconstruction and single image defogging.

An analysis of single image defogging methods using a color ellipsoid framework

The goal of this article is to explain how several single image defogging methods work using a color ellipsoid framework. The foundation of the framework is the atmospheric dichromatic model which is analogous to the reflectance dichromatic model. A key step in single image defogging is the ability to estimate relative depth. Therefore, properties of the color ellipsoids are tied to depth cues within an image. This framework is then extended using a Gaussian mixture model to account for multiple mixtures which gives intuition in more complex observation windows, such as observations at depth discontinuities which is a common problem in single image defogging. A few single image defogging methods are analyzed within this framework and surprisingly tied together with a common approach in using a dark prior. A new single image defogging method based on the color ellipsoid framework is introduced and compared to existing methods.