Dark Channel Prior Research Articles

A novel dual-fusion algorithm of single image dehazing based on anisotropic diffusion and Gaussian filter

Dark channel prior (DCP) is a widely used method in single image dehazing technology. Here, we propose a novel dual-fusion algorithm of single image dehazing based on anisotropic diffusion and Gaussian filter to suppress the halo effect or colour distortion in traditional DCP algorithms. Anisotropic diffusion is used to edge-preserving smooth images and a Gaussian filter is to smooth the local white objects. A dual-fusion strategy is conducted to optimise the atmospheric veil. Besides, the fast explicit diffusion (FED) scheme is used to accelerate the numerical solution of the anisotropic diffusion to reduce time consumption. The subjective and objective evaluation of the experiment shows that the proposed algorithm can effectively suppress the halo effect and colour distortion, and has good dehazing performance on evaluation metrics. The proposed algorithm also reduces the time consumption by 54.2% than DCP with guided filter. This study provides an effective solution for single image dehazing.

Underwater Image Enhancement Using Customized CLAHE and Adaptive Color Correction

Underwater images degraded due to low contrast and visibility issues. Therefore, it is important to enhance the images and videos taken in the underwater environment before processing. Enhancement is a way to improve or increase image quality and to improve the contrast of degraded images. The original image or video which is captured through image processing devices needs to improve as there are various issues such as less light available, low resolution, and blurriness in underwater images caused by the normal camera. Various researchers have proposed different solutions to overcome these problems. Dark channel prior (DCP) is one of the most used techniques which produced a better Peak Signal to Noise Ratio (PSNR) value. However, DCP has some issues such as it tends to darken images, reduce contrast, and produce halo effects. The proposed method solves these issues with the help of contrast-limited adaptive histogram equalization (CLAHE) and the Adaptive Color Correction Method. The proposed method was assessed using Japan Agency for Marine-Earth Science and Technology (JAMSTEC), and some images were collected from the internet. The measure of entropy (MOE), Measure of Enhancement (EME), Mean Square Error (MSE), and PSNR opted as performance measures during experiments. The values of MSE and PSNR achieved by the proposed framework are 0.26 and 32 respectively which shows better results.

Semantic Segmentation and Inpainting of Dust with the S-Dust Dataset

In agriculture the performance of camera-based vision systems is strongly affected by environmental factors, such as rain, fog, and dust. This paper presents the first open dataset for training and evaluating image processing algorithms that remove swirling dust from images. It includes images from agricultural fields during tillage, which raises dust that is manually labeled in the images. However, dust is not a classical object that can be delineated based on edges. Moreover, the dust density is not uniformly distributed but varies and is concentrated in the area behind the implement. Therefore, different segmentation approaches were investigated. First, traditional and deep-learning methods for dust segmentation in images were compared. For the neural network, a Unet with a pretrained VGG encoder was chosen. The results show that the network cannot distinguish from the background in areas with low dust density. In contrast, the traditional segmentation method based on dark channel prior (DCP) outperforms it. Subsequently, various inpainting methods to remove dust were investigated. One observation was that dust removal from images cannot be solved with current inpainting methods. The dataset is available at *.*www.agriscapes-dataset.com

Underwater Image Restoration Based on Classification and Dark Channel Prior with Minimum Convolutional Area

Underwater Image Restoration Based on Classification and Dark Channel Prior with Minimum Convolutional Area

CloudViT: A Lightweight Vision Transformer Network for Remote Sensing Cloud Detection

Clouds inevitably exist in satellite images, which limit the processing and application of satellite images to a certain extent. Therefore, cloud detection is a preprocessing task in satellite image extraction and analysis processing. However, the existing methods are difficult to mine robust features, and the number of parameters and computation are large, which is not conducive to the deployment of the model. In this letter, cloud vision transformer (CloudViT), a lightweight vision transformer network for cloud detection from satellite imagery, is proposed. In detail, to utilize dark channel priors in multispectral imagery to guide the network to learn features, a multiscale dark channel extractor is used to first predict dark channels, and then, the dark channel features and image features are input to the attention mechanism-based dark channel-guided context aggregation module to enhance image features, which in turn makes cloud detection results more accurate. At the same time, to enhance the transfer ability of the network between different satellite sensors, a plug-and-play channel adaptive module is proposed to deal with the inconsistency of the number of different satellite sensor bands. The experimental results on the Landsat7 dataset show that our network CloudViT outperforms the state-of-the-art methods while keeping the number of parameters and computation small. At the same time, the experimental results on transfer to three other datasets show that using the channel adaptation module can greatly improve the transfer ability of the model.

Image Enhancement in Foggy Images using Dark Channel Prior and Guided Filter

Haze is very apparent in images shot during periods of bad weather (fog). The image's clarity and readability are both diminished as a result. As part of this work, we suggest a method for improving the quality of the hazy image and for identifying any objects hidden inside it. To address this, we use the picture enhancement techniques of Dark Channel Prior and Guided Filter. The Saliency map is then used to segment the improved image and identify passing vehicles. Lastly, we describe our method for calculating the actual distance in units from a camera-equipped vehicle of an item (another vehicle).Our proposed solution can warn the driver based on the distance to help them prevent an accident. Our suggested technology improves images and accurately detects vehicles nearly 100% of the time.

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.

Real-time detection of underwater river crab based on multi-scale pyramid fusion image enhancement and MobileCenterNet model

To achieve rapid and accurate detection of pond cultured river crab, an underwater river crab target detection method based on multi-scale pyramid fusion image enhancement and the MobileCenterNet model is proposed in this paper. Firstly, considering the characteristics of underwater crab image, such as blurred and uneven brightness, multi-scale pyramid fusion is proposed to enhance underwater crab image based on Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance contrast and Underwater Dark Channel Prior (UDCP) for fog removal. Secondly, a crab target detection method based on the MobileCenterNet model is presented. The improved MobileNetv2 backbone network with coordinate attention module is applied to extract crab features, which not only achieves lightweight but also focuses the model's attention on crab-related features. Then, the Feature Fusion Module (FFM) is designed to extract multi-scale feature map information, and Atrous Spatial Pyramid Pooling (ASPP) is added to fuse context information from different receptive fields. The experimental results show that the average precision (AP) and F1 values of MobileCenterNet are 97.86 % and 97.94 %, the model size is only 24.46 M, and the detection speed is reaching 48.18 frames/s. Compared with the baseline model ResNet18-CenterNet, the storage memory required for model training is reduced by 81 %, and the AP on the crab dataset is increased by 3.2 %. The experimental results show that the proposed method can achieve real-time and accurate detection of underwater river crabs, and provide effective guidance for real-time monitoring and scientific feeding in crab breeding.

LRINet: Long-range imaging using multispectral fusion of RGB and NIR images

When imaging at a long distance by ordinary visible cameras, the wavelength of visible light is easily interfered by fog or atmospheric effects, resulting in blurry or lost details in RGB images. However, NIR cameras are robust to the long distance imaging without color. In this paper, we propose long-range imaging using multispectral fusion of RGB and NIR images, called LRINet. We adopt unsupervised learning for the fusion to solve the absence of ground truth. LRINet is an end-to-end network based on convolutional neural networks (CNNs) that consists of three steps for the multispectral fusion: feature extraction, fusion and reconstruction. To align unpaired data and treat the discrepancy between RGB and NIR images, we construct WarpingNet to warp NIR features, and add it into the feature extraction. Since dark channel prior (DCP) provides distance from the camera by light transmission degree, we combine it with structure loss as the weight for fusion. To ensure the color fidelity, LRINet operates the fusion on the input of the RGB luma channel and NIR image. Experimental results show that LRINet produces natural-looking color images with clear details by successfully treating discrepancy between RGB and NIR images and outperforms state-of-art fusion models in term of both visual quality and quantitative measurements.

Traffic image haze removal based on optimized retinex model and dark channel prior

GPS monitoring systems and the development of driverless vehicles are almost inseparable from camera images. The images taken by traffic cameras often contain certain sky areas and noise, the traditional dark channel prior (DCP) algorithm easily produces color distortion and halo effect, when processing the hazy traffic images with sky and high brightness areas. An optimized Retinex model and dark channel prior algorithm (ORDCP) is proposed in this paper. Firstly by adjusting the calculation method of dark channel image, the proportion of dark channel is improved; Then, the transmittance image is corrected and smoothed by guided filtering and mean filtering. Finally, the Retinex model is fused to save the details.ORDCP corrects the inaccurate calculation of scene transmittance value in DCP algorithm,and modifies some dehazing problems, such as the loss of details, halo effect, contrast and color distortion,etc. Using information entropy (IE) as the objective evaluation index, combined with the subjective evaluation, it is concluded that the algorithm proposed in this paper can effectively retain the detailed information of the image, and eliminate the halo effect. Meanwhile, it meets the visual characteristics of human eyes better, and has some practicality and applicability in traffic control and intelligent detection.

Image enhancement for underwater range-gated image with multi-slice integration method

Active illumination underwater brings the backscatter problem. The traditional pulsed laser range-gated imaging (PLRGI) system uses a small gate width to suppress most of the backscattered light. As a result, the depth of field (DOF) of the PLRGI system is small. In addition, the DOF has to be extended when the system is used for searching the target. There are several methods to extend the DOF. However, every method has its shortcoming. To overcome these problems, a multi-slice integration (MSI) method is proposed. In turbid water, the image captured by the MSI method also has the backscatter problem. The rolling ball and intensity transformation are proposed as an image enhancement method to remove the backscattered light and improve the brightness of the image. Experiments carried out in a towing boat tank and Mulan Lake show that the proposed method is better than lower-upper-threshold correlation, dark channel priors, and unsharp filtering for backscattered light removal.

Estimation of chlorophyll distribution in banana canopy based on RGB-NIR image correction for uneven illumination

Chlorophyll generally measured by Soil and Plant Analysis Development (SPAD) meter is an important indicator for the evaluation of plant photosynthesis ability and growth status. In view of the uneven distribution of illumination on broad-leaved banana canopy in the field, a correction method based on dark channel prior (DCP) is proposed to enhance the image quality, and improve the estimating accuracy of chlorophyll content. First, RGB (Red, Green, Blue) and NIR (near-infrared band) images of banana plant were separately obtained, and the two images were matched and aligned after preprocessing. Second, the channels in the region of interest (ROI) were split and restructured. The restructured image was corrected based on the DCP principle. Then, the optimal channels were selected according to the correlation analysis to form a new RGB (NewRGB) image, to simplify that OriRGB-NIR and NewRGB-NIR indicated the multispectral image before and after correction. The image features related to color features (CFs) and vegetation indices (VIs) were extracted and optimized as input variables for the chlorophyll estimation modeling. Finally, the random forest regression (RFR) was used to build the SPAD value estimation models, in which OriRGB-NIR-RFR model and the NewRGB-NIR-RFR model were named to show the results before and after image correction. Results show that the proposed correction method of uneven illumination reduced the image contrast and obtained more uniform values. At the same time, the accuracy of the model after correction was higher than that prior to correction. The modeling Rv2 of NewRGB-NIR-RFR with 11 variables reached 0.692. As a result, the visual distribution map of chlorophyll SPAD values was established, thereby providing the basis for the dynamic monitoring of crop nutrition.

Images Enhancement of Ancient Mural Painting of Bey's Palace Constantine, Algeria and Lacuna Extraction Using Mahalanobis Distance Classification Approach.

As a result of human activity and environmental changes, several types of damages may occur to ancient mural paintings; indeed, lacunae, which refer to the area of paint layer loss, are the most prevalent kind. The presence of lacuna is an essential sign of the progress of mural painting deterioration. Most studies have focused on detecting and removing cracks from old paintings. However, lacuna extraction has not received the necessary consideration and is not well-explored. Furthermore, most recent studies have focused on using deep learning for mural protection and restoration, but deep learning requires a large amount of data and computational resources which is not always available in heritage institutions. In this paper, we present an efficient method to automatically extract lacunae and map deterioration from RGB images of ancient mural paintings of Bey’s Palace in Algeria. Firstly, a preprocessing was applied using Dark Channel Prior (DCP) to enhance the quality and improve visibility of the murals. Secondly, a determination of the training sample and pixel’s grouping was assigned to their closest sample based on Mahalanobis Distance (MD) by calculating both the mean and variance of the classes in three bands (R, G, and B), in addition to the covariance matrix of all the classes to achieve lacuna extraction of the murals. Finally, the accuracy of extraction was calculated. The experimental results showed that the proposed method can achieve a conspicuously high accuracy of 94.33% in extracting lacunae from ancient mural paintings, thus supporting the work of a specialist in heritage institutions in terms of the time- and cost-consuming documentation process.

A novel technique for spatiotemporal dahazing of video image

Recently video surveillance in smart city projects is becoming more and more popular. Generally, high quality image is required in video image analysis and recognition. Often bad weather conditions like atmospheric haze, fog, and smoke affect captured outdoor images and result in loss of visibility and poor contrast. In this paper, we propose a new method for a single image and video dehazing. Many complex methods are existing for removing haze from hazy images. In this paper, we propose a method that combines dark channel prior(DCP) and bright channel prior(BCP) along with a guided filtering technique to perform effectively and efficiently by spatiotemporal means in video dehazing. To extract the global atmospheric light accurately, we exploit multiple prior DCP and BCP underlying hazy images. In addition, the rough transmission map is estimated and improvised using a guided filter to get refined transmission. The experimental result shows that our proposed algorithm enhances the colour fidelity reduces the halo effect and improves the efficiency of video dehazing.

Single-shot retinal image enhancement using untrained and pretrained neural networks priors integrated with analytical image priors

Retinal images acquired using fundus cameras are often visually blurred due to imperfect imaging conditions, refractive medium turbidity, and motion blur. In addition, ocular diseases such as the presence of cataracts also result in blurred retinal images. The presence of blur in retinal fundus images reduces the effectiveness of the diagnosis process of an expert ophthalmologist or a computer-aided detection/diagnosis system. In this paper, we put forward a single-shot deep image prior (DIP)-based approach for retinal image enhancement. Unlike typical deep learning-based approaches, our method does not require any training data. Instead, our DIP-based method can learn the underlying image prior while using a single degraded image. To perform retinal image enhancement, we frame it as a layer decomposition problem and investigate the use of two well-known analytical priors, i.e., dark channel prior (DCP) and bright channel prior (BCP) for atmospheric light estimation. We show that both the untrained neural networks and the pretrained neural networks can be used to generate an enhanced image while using only a single degraded image. The proposed approach is time and memory-efficient, which makes the solution feasible for real-world resource-constrained environments. We evaluate our proposed framework quantitatively on five datasets using three widely used metrics and complement that with a subjective qualitative assessment of the enhancement by two expert ophthalmologists. For instance, our method has achieved significant performance for untrained CDIPs coupled with DCP in terms of average PSNR, SSIM, and BRISQUE values of 40.41, 0.97, and 34.2, respectively, and for untrained CDIPs coupled with BCP, it achieved average PSNR, SSIM, and BRISQUE values of 40.22, 0.98, and 36.38, respectively. Our extensive experimental comparison with several competitive baselines on public and non-public proprietary datasets validates the proposed ideas and framework.

GAN-Based Image Dehazing for Intelligent Weld Shape Classification and Tracing Using Deep Learning

Weld seam identification with industrial robots is a difficult task since it requires manual edge recognition and traditional image processing approaches, which take time. Furthermore, noises such as arc light, weld fumes, and different backgrounds have a significant impact on traditional weld seam identification. To solve these issues, deep learning-based object detection is used to distinguish distinct weld seam shapes in the presence of weld fumes, simulating real-world industrial welding settings. Genetic algorithm-based state-of-the-art object detection models such as Scaled YOLOv4 (You Only Look Once), YOLO DarkNet, and YOLOv5 are used in this work. To support actual welding, the aforementioned architecture is trained with 2286 real weld pieces made of mild steel and aluminum plates. To improve weld detection, the welding fumes are denoised using the generative adversarial network (GAN) and compared with dark channel prior (DCP) approach. Then, to discover the distinct weld seams, a contour detection method was applied, and an artificial neural network (ANN) was used to convert the pixel values into robot coordinates. Finally, distinct weld shape coordinates are provided to the TAL BRABO manipulator for tracing the shapes recognized using an eye-to-hand robotic camera setup. Peak signal-to-noise ratio, the structural similarity index, mean square error, and the naturalness image quality evaluator score are the dehazing metrics utilized for evaluation. For each test scenario, detection parameters such as precision, recall, mean average precision (mAP), loss, and inference speed values are compared. Weld shapes are recognized with 95% accuracy using YOLOv5 in both normal and post-fume removal settings. It was observed that the robot is able to trace the weld seam more precisely.

GUDCP: Generalization of Underwater Dark Channel Prior for Underwater Image Restoration

This letter introduces an underwater image enhancement method to handle low contrast and color cast of underwater images. Firstly, with the help of hierarchical searching technique, we propose a novel backscattered light estimation method. And in this procedure, a novel scoring formula is considered into our method, which comprehensively considers multiple prior knowledge. Then, we generalize underwater dark channel prior (UDCP) approach to obtain more robust transmission estimation. In addition, we also develop a white balance method to further modify the appearance of the resultant image. Extensive experiments on real-world images demonstrate that the proposed method outperforms several previous image restoration or enhancement works.

Single Image Dehazing Based on Two-Stream Convolutional Neural Network

Objective The haze weather environment leads to the deterioration of the visual effect of the image, and it is difficult to carry out the work of the advanced vision task. Therefore, dehazing the haze image is an important step before the execution of the advanced vision task. Traditional dehazing algorithms achieve image dehazing by improving image brightness and contrast or constructing artificial priors such as color attenuation priors and dark channel priors, but the effect is unstable when dealing with complex scenes. In the method based on convolutional neural network, the image dehazing network of the encoding and decoding structure does not consider the difference before and after the dehazing image, and the image spatial information is lost in the encoding stage. In order to overcome these problems, this paper proposes a novel end-to-end two-stream convolutional neural network for single image dehazing. Method The network model is composed of a spatial information feature stream and a high-level semantic feature stream. The spatial information feature stream retains the detailed information of the dehazing image, and the high-level semantic feature stream extracts the multi-scale structural features of the dehazing image. A spatial information auxiliary module is designed between the feature streams. This module uses the attention mechanism to construct a unified expression of different types of information, and realizes the gradual restoration of the clear image with the semantic information auxiliary spatial information in the dehazing network. A parallel residual twicing module is proposed, which performs dehazing on the difference information of features at different stages to improve the model’s ability to discriminate haze images. Result The peak signal-to-noise ratio and structural similarity are used to quantitatively evaluate the similarity between the dehazing results of each algorithm and the original image. The structure similarity and peak signal-to-noise ratio of the method in this paper reached 0.852 and 17.557dB on the Hazerd dataset, which were higher than all comparison algorithms. On the SOTS dataset, the indicators are 0.955 and 27.348dB, which are sub-optimal results. In experiments with real haze images, this method can also achieve excellent visual restoration effects.

Eximious Sandstorm Image Improvement Using Image Adaptive Ratio and Brightness-Adaptive Dark Channel Prior

Sandstorm images have a color cast by sand particles. Hazy images have similar features to sandstorm images due to these images having a common obtaining process. To improve hazy images, various dehazing methods are being studied. However, not all methods are appropriate for enhancing sandstorm images as they experience color degradation via an imbalanced color channel and degraded color distributed around the image. Therefore, this paper proposes two steps to improve sandstorm images. The first is a color-balancing step using the mean ratio of the color channel between red and other colors. The sandstorm image has a degraded color channel, and therefore, the attenuated color channel has different average values for each color channel; the red channel’s average value is the highest, and that of the blue channel is the lowest. Using this property, this paper balances the color of images via the ratio of color channels. Although the image is enhanced, if the red channel is still the most abundant, the enhanced image may have a reddish color. Therefore, to enhance the image naturally, the red channel is adjusted by the average ratio of the color channel; those measures (as with the average ratio of color channels) are called image adaptive ratio (IAR). Because color-balanced sandstorm images have the same characteristics as hazy images, to enhance them, a dehazing method is applied. Ordinary dehazing methods often use dark channel prior (DCP). Though DCP estimates the dark region of an image, because the intensity of brightness is too high, the estimated DCP is not sufficiently dark. Additionally, DCP is able to show the artificial color shift in the enhanced image. To compensate for this point, this paper proposes a brightness-adaptive dark channel prior (BADCP) using a normalized color channel. The image improved using the proposed method has no color distortion or artificial color. The experimental results show the superior performance of the proposed method in comparison with state-of-the-art dehazing methods, both subjectively and objectively.

Marine Application Evaluation of Monocular SLAM for Underwater Robots.

With the development of artificial intelligence technology, visual simultaneous localization and mapping (SLAM) has become a cheap and efficient localization method for underwater robots. However, there are many problems in underwater visual SLAM, such as more serious underwater imaging distortion, more underwater noise, and unclear details. In this paper, we study these two problems and chooses the ORB-SLAM2 algorithm as the method to obtain the motion trajectory of the underwater robot. The causes of radial distortion and tangential distortion of underwater cameras are analyzed, a distortion correction model is constructed, and five distortion correction coefficients are obtained through pool experiments. Comparing the performances of contrast-limited adaptive histogram equalization (CLAHE), median filtering (MF), and dark channel prior (DCP) image enhancement methods in underwater SLAM, it is found that the DCP method has the best image effect evaluation, the largest number of oriented fast and rotated brief (ORB) feature matching, and the highest localization trajectory accuracy. The results show that the ORB-SLAM2 algorithm can effectively locate the underwater robot, and the correct distortion correction coefficient and DCP improve the stability and accuracy of the ORB-SLAM2 algorithm.