Hazy Input Research Articles

SelfPromer: Self-Prompt Dehazing Transformers with Depth-Consistency

This work presents an effective depth-consistency Self-Prompt Transformer, terms as SelfPromer, for image dehazing. It is motivated by an observation that the estimated depths of an image with haze residuals and its clear counterpart vary. Enforcing the depth consistency of dehazed images with clear ones, therefore, is essential for dehazing. For this purpose, we develop a prompt based on the features of depth differences between the hazy input images and corresponding clear counterparts that can guide dehazing models for better restoration. Specifically, we first apply deep features extracted from the input images to the depth difference features for generating the prompt that contains the haze residual information in the input. Then we propose a prompt embedding module that is designed to perceive the haze residuals, by linearly adding the prompt to the deep features. Further, we develop an effective prompt attention module to pay more attention to haze residuals for better removal. By incorporating the prompt, prompt embedding, and prompt attention into an encoder-decoder network based on VQGAN, we can achieve better perception quality. As the depths of clear images are not available at inference, and the dehazed images with one-time feed-forward execution may still contain a portion of haze residuals, we propose a new continuous self-prompt inference that can iteratively correct the dehazing model towards better haze-free image generation. Extensive experiments show that our SelfPromer performs favorably against the state-of-the-art approaches on both synthetic and real-world datasets in terms of perception metrics including NIQE, PI, and PIQE. The source codes will be made available at https://github.com/supersupercong/SelfPromer.

Weighted Fusion of Pre-processing Techniques for Neural Network-based Image Haze Removal

Haze is the natural phenomenon, which affects an image's air light and visibility. It creates a layer that hides the information in an acquired hazy image and decreases its visibility. Hazy scenarios are mostly seen in the transportation sector and remote sensing. It affects the quality of an image captured. Haze is one of the major hurdles in several computer vision applications. This paper observes and analyses different methods of haze removal via image enhancement techniques. Proposes the weighted average of the image enhancement methods to generate the enhanced hazy input image as the initial step. These enhanced images do train the neural network to estimate transmission map as well as atmospheric light, used for haze removal from images. The proposed method is experimented with 135 hazy images from three standard datasets, alias I-Haze, NH-Haze, and O-Haze (45 images from each total 135 hazy images). It gives clearer results than a few similar existing haze removal techniques. Also, the experimental results tested with performance metrics Entropy PSNR, and SSIM have demonstrated the effectiveness of the proposed haze removal method having weighted fusion of pre-processing techniques.

DPAFD-net: A dual-path adaptive fusion dehazing network

Image dehazing is an ill-posed problem that has been extensively studied in recent years. Unfortunately, most existing deep dehazing models have high computational complexity and lack the dynamic adjustment of details, which hinders their application to high-resolution images in computational vision tasks. In this paper, we propose an efficient dual-path adaptive fusion dehazing network (DPAFD-Net) to directly restore a clear image from a hazy input. Moreover, we propose a pure subnetwork with encoder and decoder structures to further extract the structural information and progressively restore the haze-free image. To evaluate the effectiveness of the proposed method, we validate our approach on synthetic and real hazy images, where our method performs favourably against the state-of-the-art dehazing approaches.

WMCP-EM: An integrated dehazing framework for visibility restoration in single image

Adopting local patches for varying haze conditions is crucial for optimizing the performance in single image dehazing. We propose a novel two-fold method with a prime focus on self-adaptive prior named Weighted Median Channel Prior (WMCP) to resolve the problems introduced by using a fixed size local-patch in the dehazing process. WMCP works by leveraging the spatially changing haze statistics such as inclusion–exclusion of related pixels for estimating depth-map in varying haze conditions. It is a scale-invariant technique that retains most of the information present in the local neighbourhood of the hazy input image for estimating scene depth, which traditional methods generally fail to preserve. In addition, an unsharp-masking based technique called edge-modulation (EM) promotes hidden or missing details such as microscopic edges and textures lost due to haze, making this scheme beneficial in ensuring a visually aesthetic and realistic dehazed image. This research also includes a set of ablation tests to assess the contributions of each module engaged in the dehazing process. We performed a comparative evaluation of our method with several state-of-the-art techniques, revealing its superiority in terms of visibility improvement and edge preservation, especially when the dense haze regions are taken into consideration.

Single-image Dehazing using Detail Enhancement and Image Fusion

Haze is the suspension of atmospheric particles which is sufficient to reduce the visibility. Image dehazing refers to the processing tasks that lessen this negative effect. In this work, an alternative approach to single-image dehazing is developed which skips solving the haze formation equation, while still respects its hypothesis. In this method, we generated multiple under-exposed images from a single hazy input, followed by a detail enhancement process. Such resulting images were then merged using weights calculated based on the Dark Channel Prior assumption and overcame luminance enhancement. The visual improvement has been validated by both qualitative and quantitative evaluations.

Nighttime Image Dehazing Based on Multi-Scale Gated Fusion Network

In this paper, we propose an efficient algorithm to directly restore a clear image from a hazy input, which can be adapted for nighttime image dehazing. The proposed algorithm hinges on a trainable neural network realized in an encoder–decoder architecture. The encoder is exploited to capture the context of the derived input images, while the decoder is employed to estimate the contribution of each input to the final dehazed result using the learned representations attributed to the encoder. The constructed network adopts a novel fusion-based strategy which derives three inputs from an original input by applying white balance (WB), contrast enhancing (CE), and gamma correction (GC). We compute pixel-wise confidence maps based on the appearance differences between these different inputs to blend the information of the derived inputs and preserve the regions with pleasant visibility. The final clear image is generated by gating the important features of the derived inputs. To train the network, we introduce a multi-scale approach to avoid the halo artifacts. Extensive experimental results on both synthetic and real-world images demonstrate that the proposed algorithm performs favorably against the state-of-the-art dehazing for nighttime images.

Zero-Shot Remote Sensing Image Dehazing Based on a Re-Degradation Haze Imaging Model

Image dehazing is crucial for improving the advanced applications on remote sensing (RS) images. However, collecting paired RS images to train the deep neural networks (DNNs) is scarcely available, and the synthetic datasets may suffer from domain-shift issues. In this paper, we propose a zero-shot RS image dehazing method based on a re-degradation haze imaging model, which directly restores the haze-free image from a single hazy image. Based on layer disentanglement, we design a dehazing framework consisting of three joint sub-modules to disentangle the hazy input image into three components: the atmospheric light, the transmission map, and the recovered haze-free image. We then generate a re-degraded hazy image by mixing up the hazy input image and the recovered haze-free image. By the proposed re-degradation haze imaging model, we theoretically demonstrate that the hazy input and the re-degraded hazy image follow a similar haze imaging model. This finding helps us to train the dehazing network in a zero-shot manner. The dehazing network is optimized to generate outputs that satisfy the relationship between the hazy input image and the re-degraded hazy image in the re-degradation haze imaging model. Therefore, given a hazy RS image, the dehazing network directly infers the haze-free image by minimizing a specific loss function. Using uniform hazy datasets, non-uniform hazy datasets, and real-world hazy images, we conducted comprehensive experiments to show that our method outperforms many state-of-the-art (SOTA) methods in processing uniform or slight/moderate non-uniform RS hazy images. In addition, evaluation on a high-level vision task (RS image road extraction) further demonstrates the effectiveness and promising performance of the proposed zero-shot dehazing method.

Feature Attention Parallel Aggregation Network for Single Image Haze Removal

Images captured in hazy weather often suffer from color distortion and texture blur due to turbid media suspended in the atmosphere. In this paper, we propose a Feature Attention Parallel Aggregation Network (FAPANet) to restore a clear image directly from the corresponding hazy input. It adopts the encoder-decoder structure while incorporating residual learning and attention mechanism. FAPANet consists of two key modules: a novel feature attention aggregation module (FAAM) and an adaptive feature fusion module (AFFM). FAAM recalibrates features by integrating channel attention and pixel attention in parallel to stimulate useful information and suppress redundant features. The shallow and deep layers of neural networks tend to characterize the low-level and high-level semantic features of images, respectively, so we introduce AFFM to fuse these two features adaptively. Meanwhile, a joint loss function, composed of L1 loss, perceptual loss, and structural similarity (SSIM) loss, is employed in the training stage for better results with more vivid colors and richer details. Comprehensive experiments on both synthetic and real-world images demonstrate the impressive performance of the proposed approach.

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.

IDRLP: Image Dehazing Using Region Line Prior.

In this work, a novel and ultra-robust single image dehazing method called IDRLP is proposed. It is observed that when an image is divided into n regions, with each region having a similar scene depth, the brightness of both the hazy image and its haze-free correspondence are positively related with the scene depth. Based on this observation, this work determines that the hazy input and its haze-free correspondence exhibit a quasi-linear relationship after performing this region segmentation, which is named as region line prior (RLP). By combining RLP and the atmospheric scattering model (ASM), a recovery formula (RF) can be easily obtained with only two unknown parameters, i.e., the slope of the linear function and the atmospheric light. A 2D joint optimization function considering two constraints is then designed to seek the solution of RF. Unlike other comparable works, this "joint optimization" strategy makes efficient use of the information across the entire image, leading to more accurate results with ultra-high robustness. Finally, a guided filter is introduced in RF to eliminate the adverse interference caused by the region segmentation. The proposed RLP and IDRLP are evaluated from various perspectives and compared with related state-of-the-art techniques. Extensive analysis verifies the superiority of IDRLP over state-of-the-art image dehazing techniques in terms of both the recovery quality and efficiency. A software release is available at https://sites.google.com/site/renwenqi888/.

CSIDNet: Compact single image dehazing network for outdoor scene enhancement

This paper proposes a novel deep learning-based single image dehazing network named as Compact Single Image Dehazing Network (CSIDNet) for outdoor scene enhancement. CSIDNet directly outputs a haze-free image from the given hazy input. The remarkable features of CSIDNet are that it has been designed only with three convolutional layers and it requires lesser number of images for training without diminishing the performance in comparison to the other commonly observed deep learning-based dehazing models. The performance of CSIDNet has been analyzed on natural hazy scene images and REalistic Single Image DEhazing (RESIDE) dataset. RESIDE dataset consists of Outdoor Training Set (OTS), Synthetic Objective Testing Set (SOTS), and real-world synthetic hazy images from Hybrid Subjective Testing Set (HSTS). The performance metrics used for comparison are Peak Signal to Noise Ratio (PSNR) and Structural SIMilarity (SSIM) index. The experimental results obtained using CSIDNet outperform several well known state-of-the-art dehazing methods in terms of PSNR and SSIM on images of SOTS and HSTS from RESIDE dataset. Additionally, the visual comparison shows that the dehazed images obtained using CSIDNet are more appealing with better edge preservation. Since the proposed network requires minimal resources and is faster to train along with lesser run-time, it is more practical and feasible for real-time applications.

An improvement of visual perception of single image by fusion of DCP with patch processing

Scattering of atmospheric particles in hazy or foggy weather conditions is the primary reason for the degradation of images. Though a level of saturation has been obtained in the improvement of restoration model to overcome the visual vividness those termed as a dehazing method, there still remains a void while restoring of the digital image. We, in this paper, have proposed a novel and efficient haze removal algorithm by considering an efficient prior knowledge like DCP in order to eliminate haze and, restore albedo by combined with diffusion patch process after soft mating to enhance again for better clarity and accuracy of a hazy input image. In this research analysis, we critically examined popular Dark Channel Prior haze resisting removals algorithm such as Fattal, He, Tan and Tarel with our proposed novel haze removal model, The image quality assessment such as qualitative and quantitative index of the de-hazed image is concerned with estimating the parameters such as RSME, PSNR, SSIM, FSIM, and FSIMc are also reported for measuring precise accuracy of our proposed methodology as compared to the existing one. This novel method found to be distinctive and ensures efficient production of a high-quality haze-free image.

Image Haze Removal via Reference Retrieval and Scene Prior.

Photography of hazy scene typically suffers from low-contrast which degrades the visibility of the scene. The performance of single-image dehazing methods is limited by the priors or constraints. In this paper, we present an effective method for haze removal, which utilizes its retrieved correlated haze-free images as external information. The correlated haze-free images are with scene prior offering scene structure and local high frequency information for dehazing, although variations in viewpoints, scales, and illumination conditions exist. To utilize those reference more effectively, global geometric registration and local block matching toward the hazy input are performed to reinforce the spatial correlations. Based on the registration, different kinds of external information are estimated. In addition, we combine that additional external information with internal constraint and regularization for estimating scene transmission map. Experiments demonstrate that our approach can produce dehazing results with better visual quality compared with other state-of-the-art methods.

Fusion-based Variational Image Dehazing

We propose a novel image-dehazing technique based on the minimization of two energy functionals and a fusion scheme to combine the output of both optimizations. The proposed fusion-based variational image-dehazing (FVID) method is a spatially varying image enhancement process that first minimizes a previously proposed variational formulation that maximizes contrast and saturation on the hazy input. The iterates produced by this minimization are kept, and a second energy that shrinks faster intensity values of well-contrasted regions is minimized, allowing to generate a set of difference-of-saturation (DiffSat) maps by observing the shrinking rate. The iterates produced in the first minimization are then fused with these DiffSat maps to produce a haze-free version of the degraded input. The FVID method does not rely on a physical model from which to estimate a depth map, nor it needs a training stage on a database of human-labeled examples. Experimental results on a wide set of hazy images demonstrate that FVID better preserves the image structure on nearby regions that are less affected by fog, and it is successfully compared with other current methods in the task of removing haze degradation from faraway regions.

A GPU-accelerated real-time single image de-hazing method using pixel-level optimal de-hazing criterion

Single image de-hazing is an important and challenging research topic in computer vision. The computational efficiency and robustness of this issue are key problems in real-time applications. In this paper, a graphic processing unit (GPU)-accelerated real-time image enhancing method is proposed to remove haze from a single hazy input image. The foundation of this method is a novel pixel-level optimal de-hazing criterion proposed to combine a virtual hazy-free candidate image sequence into a final single hazy-free image. This image sequence is estimated from the input hazy image by exhausting all possible discretely sampled scene depth values. The main advantage of proposed method is the single pixel independently computing fashion. Its computing at one single pixel position is independent of others. Based on this property, it is straightforward to implement the proposed method with fully parallel GPU acceleration. The sufficient experimentations on various scenes indicate that the proposed method can process a large hazy image with one megapixel to visually moderate haze-free result at a rate of 80 frames/s. Moreover, the proposed method is also less affected by the nonuniform illumination compared to previous methods.