Hardware Implementation for Haze Removal With Adaptive Filtering

Outdoor scenes are usually affected by air particles and water vapor in the atmosphere. Haze is such phenomena happened due to atmospheric absorption and scattering. Haze affects the visibility and causes a degradation in the images quality of outdoor scenes, which in turn affects various image processing applications. In this paper, a review of the recent haze removal methods and its applications is introduced. The haze removal methods are categorized according to the input of their algorithm into two categories: multiple images and single image. Afterward, the different strategies in each category of the haze removal techniques are discussed and their characteristics are illustrated. Eventually, the quantitative parameters that evaluate the quality of the haze removed image are presented. This review paper highlights the importance of the haze removal methods in restoring a high-quality image relative to the quality of the original haze-free image. This haze removed image is a vital key in many computer vision applications including the visual navigation and object recognition and target detection.

Image degraded by haze is a critical aspect in today's environment while getting a high-quality haze-free image remains an important task in computer vision. In recent year, many works have been done to improve the visibility of image taken under bad weather. Conventional designs use multiple image and/ or single image to deal with haze removal. In this paper, we use a simple but effective prior, a variation of distance (VoD) prior, to estimate the transmission map and remove haze from a single input image. The VoD prior is developed based on the idea that the outdoor visibility of images taken under hazy weather conditions seriously reduced when the distance increases. The thickness of the haze can be estimated effectively and a haze-free image can be recovered by adopting the VoD prior and the new haze imaging model. Our method is stable to image local regions containing objects in different depths. Our experiments showed that the proposed method achieved better results than several state-of-the-art methods, and it can be implemented very quickly. Our method due to its fast speed and the good visual effect is suitable for real-time applications. This work confirms that estimating the transmission map using the distance information instead the color information is a crucial point in image enhancement and especially single image haze removal.

Haze removal is useful in computational photography and computer vision applications. Although many haze removal algorithms have been proposed, their computational efficiency requires improvement. A real-time haze removal method is presented in this paper. The method is based on the concept of a dark channel prior. To enhance the haze removal performance, an approximate method to estimate the atmospheric light and transmission is employed. For embedded system applications, a hardware architecture to perform real-time haze removal is proposed. The hardware can achieve 116 MHz on Stratix FPGA. The simulation results indicate that the hardware is highly efficient and performs well. It obtains good image recovery results and satisfies the real-time requirement even for large images.

As a challenging problem, image haze removal plays an important role in computer vision applications. The dark channel prior has been widely studied for haze removal since it is simple and effective; however, it still suffers from over‐saturation, artefacts and dark‐look. To resolve these problems, this study proposes a method of single image haze removal using content‐adaptive dark channel and post enhancement. The main contributions of this work are as follows: first, an associative filter, which can transfer the structures of a reference image and the grey levels of a coarse image to the filtering output, is employed to compute the dark channel efficiently and effectively. Secondly, the dark channel confidence is utilised to restrict the dark channel based on the content of the image. Finally, a post enhancement method is devised to map the luminance of the restored haze‐free image with the preservation of local contrast. Experimental results demonstrate that the proposed method significantly improves the visibility of the hazy image.

Computer vision applications in outdoor environment are mainly affected by factors such as pollution, clouds, shadow, haze, fog. Herein, an algorithm for real-time haze detection and removal in image has been implemented. For this purpose, the Dark Channel Prior technique is used which is efficient method for haze removal and also contains information about the level of haze in the image. Initially, visibility index of image frame is estimated to determine whether image is hazy or not using haze detection model. Thereafter, hazy images are processed with haze removal model to enhance the image visibility. The haze removal model uses guided filter to accelerate the process. Non-hazy images do not require to process with haze removal model. The algorithm has been tested for four datasets, i.e., non-hazy, slightly hazy, medium hazy, and heavily hazy. The proposed algorithm is performing well and also able to tackle the halo effects at some extent.

A constrained total variation model for single image dehazing

Under the weather of haze, fog, or smoke, outdoor images show poor visibility and low contrast. Low contrast results in the difficulty for carrying out basic local feature (e.g., interest points and edges) detection algorithms, which are necessary procedures in some computer vision applications. Hence, increasing contrast of degraded images is very important since it is helpful in finding more distinct features from haze images. However, few single image haze removal methods can simultaneously achieve clear visibility, sufficiently high contrast, and simplicity. In this paper, we propose an intuitive and effective method, called the depth-based contrast stretching transform (DCST), to simultaneously obtain clear visibility and enhance contrast of a single haze gray image. The DCST stretches the contrast of haze images based on the coarse depth layers of scenes. Our method is simple and almost real time and can be extended to color images. We analyze in detail that the image stretched by the DCST has a higher local contrast than the image recovered via the physical-based model. Experiments demonstrate that images stretched by the DCST have excellent visibility and contrast compared with a few existing algorithms. Compelling performance is also presented by comparing the proposed method with other representative methods in the application of local feature detection.

In this paper, we propose a simple but effective image prior - dark channel prior to remove haze from a single input image. The dark channel prior is a kind of statistics of the haze-free outdoor images. It is based on a key observation - most local patches in haze-free outdoor images contain some pixels which have very low intensities in at least one color channel. Using this prior with the haze imaging model, we can directly estimate the thickness of the haze and recover a high quality haze-free image. Results on a variety of outdoor haze images demonstrate the power of the proposed prior. Moreover, a high quality depth map can also be obtained as a by-product of haze removal.

Haze removal or dehazing has been a challenge in the field of image restoration. Recently, He et al.. proposed a single image dehazing scheme based on an interesting statistical prior called the dark channel prior (DCP). By the DCP, two parameters in the haze image model, the atmospheric light and the transmission map, can be estimated easily. Consequently, the DCP scheme has attracted much attention in this field. Note that the DCP scheme relies on the block-based dark channel which is considered as a strong DCP assumption. In this paper, a pixel-based dark channel is introduced through which the atmospheric light and the transmission map are estimated. The pixel-based dark channel is considered as a weak DCP (WDCP) since its statistical property is not as strong as that in the block-based dark channel. With a similar manner in the DCP scheme, the atmospheric light is estimated through the pixel-based dark channel. To make the pixel-based dark channel feasible in the transmission map estimation, an adaptive scaling factor for the initial transmission map is employed and the pixel-based dark channel is applied as the guide image in the transmission map refinement by the guided image filtering. Furthermore, an objective assessment is used to evaluate the proposed WDCP scheme and the compared DCP scheme. Simulation results indicate that the proposed WDCP scheme is more efficient, 24.30 times faster than the DCP scheme on average. Moreover, the proposed WDCP scheme is of better subjective visual quality than the DCP scheme and the employed objective assessment generally agrees with the results in the given examples.

Single-image haze removal is challenging due to its ill-posed nature. The breadth of real-world scenarios makes it difficult to find an optimal dehazing approach that works well for various applications. This article addresses this challenge by utilizing a novel robust quaternion neural network architecture for single-image dehazing applications. The architecture's performance to dehaze images and its impact on real applications, such as object detection, is presented. The proposed single-image dehazing network is based on an encoder-decoder architecture capable of taking advantage of quaternion image representation without interrupting the quaternion dataflow end-to-end. We achieve this by introducing a novel quaternion pixel-wise loss function and quaternion instance normalization layer. The performance of the proposed QCNN-H quaternion framework is evaluated on two synthetic datasets, two real-world datasets, and one real-world task-oriented benchmark. Extensive experiments confirm that the QCNN-H outperforms state-of-the-art haze removal procedures in visual quality and quantitative metrics. Furthermore, the evaluation shows increased accuracy and recall of state-of-the-art object detection in hazy scenes using the presented QCNN-H method. This is the first time the quaternion convolutional network has been applied to the haze removal task.

Haze is a common phenomenon often produced during bad weather that obscure scenes, reduces visibility, and degrades colors. Haze removal poses a serious challenge because of the difficulty to develop an exact mathematical model when a single hazy image is given as the input. In this paper, we propose a single image haze removal method to highlight the edges. A edge-aware constraint weighting scheme based on first-order derivative from Gradient domain Guided Image filter is adopted in this algorithm which helps to preserve the edges better. In this dehazing algorithm, simplified dark channel of the haze image can be separated as base layer and detail layer by applying a filter, i.e., weighted guided image filter. The transmission map is obtained from the base layer and thereby it is utilized to restore haze-free image. Result analysis show that the proposed algorithm preserves and highlights the edges without any color distortion. Experimental results are performed on natural haze images, aerial images, and under water images. In qualitative analysis, edge map shows the better performance of the proposed method. Quantitative Analysis have been carried out which demonstrates the better performance of the proposed algorithm when compared with the existing method.

Existing haze removal approaches have unreliable estimation of atmospheric light and transmission, which lead to imprecision color information recovery, gradient reversal artifacts, and halo effect. In this paper, a novel haze removal technique based on ambient light model is presented. Our technique applies canny edge detection algorithm which can locate the sky area accurately, and adopts binary-tree algorithm to obtain the radiance of atmospheric light A rapidly; then, an adaptive wiener filter is employed for modifying transmission map, and halo effect is further restricted in the sudden changed region of scene depth by the morphology method. Experimental results verify that the optimized approach has better robustness, and is superior to present methods.

Single image haze removal is crucial in computer vision. In open literatures, two kinds of dehazing strategies (prior-based and learning-based methods) have been developed. However, they have a trade-off between detail preservation and the image quality. Prior-based methods reconstruct the detail well but have lower image quality while learning-based methods achieve better recovered quality but lose the detail. In this paper, to mitigate this dilemma, a hierarchical architecture using the discrete wavelet transform (DWT) is proposed. It divides the dehazing problem into two parts: detail and background reconstruction. Based on investigating how haze affects the image in the wavelet domain, two networks for detail and background reconstruction are proposed. To avoid color distortion and the detail loss, the anti-vanish wavelet loss and the bound penalty are proposed. The multi-level wavelet component discriminator is proposed for further improvement. Experiments show that the proposed network can achieve superior performance in all metrics.

In this paper, we propose an accurate single image haze removal method that divides hazed images into regions in advance, where haze removal processing is difficult, and applies it to the estimation of ambient light. In addition, we propose a method of iterative haze removal for further improving image accuracy. The conventional method has a problem that the amount of haze is estimated larger than the actual amount in the regions where the distance between the photographic spot and subject is long, and causes that ambient light cannot be estimated accurately. The regions which are difficult to be processed mainly white and sky regions. If those regions are processed in the same way as the other regions, bad effects are observed such as the image becomes darker or its color changes too much. In the proposed method, the regions which are difficult to estimate the amount of haze are divided into multiple regions by extracting pixels similar to the feature pixels, and then ambient light is determined from the regions where haze removal processing is easy. The proposed method can easily extract the appropriate color of haze for any given image.

As a local filter, the guided image filtering (GIF) suffers from halo artifacts. To address this issue, a novel weighted aggregating strategy is proposed in this paper. By introducing the weighted aggregation to GIF, the proposed method called WAGIF can achieve results with sharp edges and avoid/reduce halo artifacts. More specifically, compared to the weighted guided image filtering and the gradient domain guided image filtering, the proposed method can achieve both fine and coarse smoothing results in the flat areas while preserving edges. In addition, the complexity of the proposed approach is O(N) for an image with N pixels. It is demonstrated that the GIF with weighted aggregation performs well in the fields of computational photography and image processing, including single image detail enhancement, tone mapping of high-dynamic-range images, single image haze removal, etc.