Dark Channel Research Articles

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.

Physical-model guided self-distillation network for single image dehazing.

Image dehazing, as a key prerequisite of high-level computer vision tasks, has gained extensive attention in recent years. Traditional model-based methods acquire dehazed images via the atmospheric scattering model, which dehazed favorably but often causes artifacts due to the error of parameter estimation. By contrast, recent model-free methods directly restore dehazed images by building an end-to-end network, which achieves better color fidelity. To improve the dehazing effect, we combine the complementary merits of these two categories and propose a physical-model guided self-distillation network for single image dehazing named PMGSDN. First, we propose a novel attention guided feature extraction block (AGFEB) and build a deep feature extraction network by it. Second, we propose three early-exit branches and embed the dark channel prior information to the network to merge the merits of model-based methods and model-free methods, and then we adopt self-distillation to transfer the features from the deeper layers (perform as teacher) to shallow early-exit branches (perform as student) to improve the dehazing effect. For I-HAZE and O-HAZE datasets, better than the other methods, the proposed method achieves the best values of PSNR and SSIM being 17.41dB, 0.813, 18.48dB, and 0.802. Moreover, for real-world images, the proposed method also obtains high quality dehazed results. Experimental results on both synthetic and real-world images demonstrate that the proposed PMGSDN can effectively dehaze images, resulting in dehazed results with clear textures and good color fidelity.

Underwater Image Enhancement Based on Color Balance and Multi-Scale Fusion

Underwater light absorption and scattering lead to color deviation, low brightness, fuzzy details and low contrast of underwater images. In this contribution, an underwater image enhancement algorithm based on color balance and multi-scale fusion is proposed. Firstly, a color balance method is used to correct the image color. Then, based on an improved dark channel prior algorithm, the local contrast information of the color balanced image is used to derive two atmospheric lights and transmission maps, which are transformed into the images with enhanced contrast and brightness, respectively. Finally, the multi-scale fusion method is adopted to fuse the contrast enhanced image and the brightness improved image according to the weights. The proposed algorithm is compared with other underwater image enhancement algorithms in qualitative and quantitative evaluation. Experimental results show that the proposed algorithm can effectively eliminate color deviation, remove dark areas, and improve brightness and contrast of underwater images. The enhanced images generated by the proposed algorithm are superior to those generated by other algorithms in indicators of PSNR, SSIM, UIQM and UCIQE. Thus, the proposed algorithm is suitable for underwater image enhancement.

KM3NeT upper bounds of detection rates of solar neutrinos from annihilations of dark matter at the solar core

The Weakly Interacting Massive Particles (WIMPs) so far remain one of the most popular candidates for dark matter. If captured gravitationally inside the core of the Sun, these WIMPs may produce high-energy neutrinos as the end product in case they undergo self-annihilations at the solar core. In this work, we address the detectability of such neutrinos at the proposed KM3NeT detector. Upper bounds of the detection rate for such neutrinos at KM3NeT are computed for the case of a generic dark matter scenario and also when specific models for particle dark matter are chosen. In this work, upper bounds of muon event rates for different annihilating dark matter masses are computed for each of the cases of dark matter annihilation channels (e.g. [Formula: see text], [Formula: see text], [Formula: see text], etc.). These upper bounds are also computed by considering the dark matter scattering cross-section using upper bounds obtained from PandaX-4T direct dark matter search experiment.

Three-stage fusion framework for single image dehazing

Fog generally causes loss or blurring of image information, significantly affecting work and life. This study proposes a three-stage fusion method comprising multiscale dark channel image fusion based on the dark channel prior, multiexposure image fusion using a gamma-correction technique, and a final fusion combining the restoration-based and enhancement-based methods to ensure good saturation and contrast in the dehazing result. The proposed method is experimentally evaluated in detail, and qualitative as well as quantitative analyses are performed. The obtained results indicate that our method can effectively improve the haze removal performance and is advantageous for dealing with long-distance hazy images.

Aerial Image Dehazing Using Reinforcement Learning

Aerial observation is usually affected by the Earth’s atmosphere, especially when haze exists. Deep reinforcement learning was used in this study for dehazing. We first developed a clear–hazy aerial image dataset addressing various types of ground; we then compared the dehazing results of some state-of-the-art methods, including the classic dark channel prior, color attenuation prior, non-local image dehazing, multi-scale convolutional neural networks, DehazeNet, and all-in-one dehazing network. We extended the most suitable method, DehazeNet, to a multi-scale form and added it into a multi-agent deep reinforcement learning network called DRL_Dehaze. DRL_Dehaze was tested on several ground types and in situations with multiple haze scales. The results show that each pixel agent can automatically select the most suitable method in multi-scale haze situations and can produce a good dehazing result. Different ground scenes may best be processed using different steps.

Fire situation detection method for unmanned fire-fighting vessel based on coordinate attention structure-based deep learning network

In the process of offshore fire protection, discovering the fire situation and grasping the development trend of the fire situation are the premise and basis for formulating scientific and reasonable firefighting strategies and programs. However, the existing fire perception algorithms were mainly applied in onshore indoor and outdoor or forest scenes, and were not able to accurately estimate fire situations in the marine environments due to the complex weather conditions and work conditions. Aiming at the fire autonomous detection problem of unmanned fire vessels at sea, we propose a fire situation perception method based on an improved dark channel prior and coordinate attention (CA)-based deep learning network for unmanned fire-fighting vessel. In the proposed model, the detected videos/images are preprocessed by the improved dark channel prior algorithm, which is used to reduce haze in the images. Additionally, an improved CA-structure is integrated with YOLOv5 framework which is a lightweight and real-time methodology so as to improve the accuracy and performance of fire detection. We term this specific setting as Dehaze-caYOLOv5. Finally, the prototype of the fire situation detection system is implemented and is able to identify different fire warning levels according to the detected results of smoke and fires. The experimental results show that the average accuracy of the model is 96.1% and the frame rate is 60fps. The method proposed in this paper is of great significance for the application of the unmanned system of offshore intelligent fire-fighting vessels.

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.

The monitoring of bridge under complex illumination based on digital image technology

In the actual bridge monitoring field, the change of illumination is an urgent problem to be solved in digital image technology. At present, when using markers to measure the displacement of bridges, problems such as the change of image gray value, uneven distribution of light, and surface defects of material cause identification difficulties. In this paper, a new recognition algorithm called the dual-channel method is proposed. Based on the dark channel principle and OSTU theory, the dual-channel method can automatically select the image gray threshold, which is used to solve the recognition problem to achieve long-term monitoring. Experimental results show this method can accurately identify the markers under complex lighting conditions and greatly expand the range of gray thresholds. Compared with the traditional displacement meter, measurement errors of the dual-channel method are less than 3%. The new method can achieve real-term effective bridge monitoring, which provides a theoretical basis for practical engineering.

Laplace dark channel attenuation-based single image defogging in ocean scenes

Laplace dark channel attenuation-based single image defogging in ocean scenes

Visibility Improvement in Air Pollution Scene by Joint Sharpness - Contrast Enhanced Dehazing

In the latter half of the 1980s, PM2.5 pollution in Beijing became a serious problem, and there were concerns about health hazards. It was expected that China's emissions must be reduced from 2013 to 2016, and the lockdown effect of Covid-19 would bring about an end, but it is still reluctant to regulate CO2 emissions. Again, in Beijing in November 2021, a visibility of 500 m or less has been observed, then road traffic is dangerous in addition to health. After that, the center of pollution has moved from India to Mongolia, and now Nepal, Qatar and Saudi Arabia. The situation is still serious in developing countries. Image restoration to remove the effects of haze and fog has been a long-standing concern of NASA, and their original Visual Servo has been put into practical use. Though the mainstream moved to the technique based on atmospheric physics. He et al.'s Dark Channel Priority (DCP) logic has had a certain effect on heavily polluted PM 2.5 scenes , but there is a limit to the restoration of detailed visibility. The observed images are affected by two spatial inhomogeneities of 1) atmospheric layer and 2) illumination. As a countermeasure, we have improved DCP process with the help of Retinex and introduced the veil coefficient as reported in CIC24. Recently, a variety of improvements in single image Dehazing, using FFA-net, BPP-net, LCA-net, or Vision-based model are in progress. However, in each case, visibility of details is still a common problem. This paper proposes an improvement in detail visibility by (1) joint sharpness-contrast preprocess and (2) adjustment in Dehaze effect with veil coefficient v. Lastly, we challenge numerical evaluation of improvement in detail visibility by the two ways of attenuation of high-frequency Fourier spectrum and the expansion rate of the color gamut.

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.

Temporal electric field of a helium plasma jet by electric field induced second harmonic (E-FISH) method

Electric field is an important parameter of plasma, which is related to electron temperature, electron density, excited species density, and so on. In this work, the electric field of an atmospheric pressure plasma jet is diagnosed by the electric field induced second harmonic (E-FISH) method, and the time-resolved electric field under different conditions is investigated. When positive pulse voltage is applied, the electric field has a peak of about 25 kV cm−1 at the rising edge of the voltage pulse. A dark channel is left behind the plasma bullet and the electric field in the dark channel is about 5 kV cm−1. On the other hand, when negative pulse voltage is applied, the electric field has a peak of −16 kV cm−1 when the negative voltage is increased to −8 kV. A relatively bright channel is left behind the plasma head and the electric field in this relatively bright channel is about −6 kV cm−1. When the pulse rising time increases from 60 to 200 ns, the peak electric field at both the rising edge and the falling edge of the voltage decreases significantly. When 0.5% of oxygen is added to the main working gas helium, the peak electric field at the rising edge is only about 15 kV cm−1. On the other hand, when 0.5% nitrogen is added, the peak electric field increases especially at the falling edge of the voltage pulse, where it increases reversely from −12 to −16 kV cm−1 (the minus sign only represents the direction of electric field).

Improved color image defogging algorithm based on dark channel prior

An improved algorithm of image defogging was proposed based on dark channel prior in order to solve the low efficiency and color distortion in the bright area using original algorithm. If the image contains large areas of bright areas such as sky, white clouds or partial white objects and water surface, we can know that the dark channel prior theory does not apply to these areas. Firstly, it is necessary to clear the bright area of the image. According to principle that he adjacent pixel attributes have similarity, the image transmittance of the local region also has similarity, Block function is Consruted. Applied the dark channel prior, judging whether each block includes a bright area by the absolute value of difference of atmospheric intensity and dark channel, the dark and bright areas of the image are obtained. So the estimation value of the adaptive space transmittance are also obtained. Secondly, the transmittance of bright region is small and it causes deviation, so the enhancement formula is used to modify it dynamically. In order to preserve the edge details after image restoration, for bright areas, using texture function to optimize transmittance independently, for others, using gradient and texture function together. Finally, it restored the fog-free image applying the atmospheric scattering model. The experimental results showed that the restored image had obvious details and rich color and fast processing speed through the proposed algorithm. The algorithm can also be applied to outdoor visual systems, such as video surveillance, intelligent traffic and so on.

Image dehazing with hybrid total variation–L0 regularization

We propose a dehazing problem model with hybrid regularization and design an effective algorithm to restore the latent image and transmission map simultaneously. In the proposed dehazing problem model, we use the total variation (TV) to regularize the latent image and adopt the hybrid TV and L0-norm (TV–L0) regularization to model the transmission map. In the proposed optimization algorithm, we first use the dark channel prior to achieve an initial guess of the global atmospheric light and transmission map. Then we convert the original problem into two subproblems: one aims to update the latent image based on TV regularization, whereas the other estimates the transmission map with hybrid TV–L0 regularization. Both of the subproblems can be solved efficiently with variable splitting and penalty technology, and the minimizer is reached by alternately solving the two subproblems. Experimental results show that our approach can achieve a high-quality restored image that is comparable to some state-of-the-art methods.

A Video-Based Real-Time Tracking Method for Multiple UAVs in Foggy Weather

Aiming at the real-time tracking problem of multiple unmanned aerial vehicles (UAVs) based on video under fog conditions, we propose a multitarget real-time tracking method that combines the Deepsort algorithm with detection based on improved dark channel defogging and improved You Only Look Once version 5 (YOLOv5) algorithm. The contributions of this paper are as follows: 1. For the multitarget tracking problem under fog interference, a multialgorithm combination method is proposed. 2. By optimizing dark channel defogging, the complexity of the original algorithm is reduced from On2 to On, which simplifies the processing time of the defogging algorithm. 3. The YOLOv5 network structure is optimized so that the network can synchronously reduce the detection time while maintaining high-precision detection. 4. The amount of algorithm processing through image size compression is reduced, and the real-time performance under high-precision tracking is improved. In the experiments conducted, the proposed method improved tracking precision by 36.1% and tracking speed by 39%. The average time of tracking per image frame was 0.036s, satisfying the real-time tracking of multiple UAVs in foggy weather.

Double-Attention YOLO: Vision Transformer Model Based on Image Processing Technology in Complex Environment of Transmission Line Connection Fittings and Rust Detection

Transmission line fittings have been exposed to complex environments for a long time. Due to the interference of haze and other environmental factors, it is often difficult for the camera to obtain high quality on-site images, and the traditional image processing technology and convolution neural networks find it difficult to effectively deal with the dense detection task of small targets with occlusion interference. Therefore, an image processing method based on an improved dark channel defogging algorithm, the fusion channel spatial attention mechanism, Vision Transformer, and the GhostNet model compression method is proposed in this paper. Based on the global receptive field of the saliency region capture and enhancement model, a small target detection network Double-attention YOLO for complex environments is constructed. The experimental results show that embedding a multi-head self-attention component into a convolutional neural network can help the model to better interpret the multi-scale global semantic information of images. In this way, the model learns more easily the distinguishable features in the image representation. Embedding an attention mechanism module can make the neural network pay more attention to the salient region of image. Dual attention fusion can balance the global and local characteristics of the model, to improve the performance of model detection.

A NEW METHOD FOR DEHAZING OF UAV REMOTE SENSING IMAGES BASED ON IMPROVED DARK CHANNEL PRIOR

Abstract. Aiming to solve the problem of loss of important image information, such as blurred details and low contrast, caused by fog, haze and other meteorological influences in the slope monitoring process of UAV remote sensing images, a new method of improving dark channel image dehazing based on channel-weighted analysis and compensation function is proposed by a training of multilayer perceptron (MLP) in this study. First, based on the dark channel prior principle, the original hazy UAV image is mapped to obtain the estimated values of atmospheric light and rough transmittance. Next, by counting the RGB three-channel values of the pixels in the high-brightness regions, and analyzing the scattering of the RGB three-channel values in the haze, a color recovery module of atmospheric light is constructed, and the estimated value of atmospheric light is optimized. Then, according to the global transmittance, the compensation boundary value is determined and a functional relationship between different brightness regions and the increment of transmittance is established as a compensation function to optimize the rough transmittance. Finally, perform secondary optimization on the rough transmittance with the multi-layer perceptron (MLP) to obtain a smoother transmittance value. The experimental results show that the image processed by the proposed method has good contrast. The color saturation and authenticity are effectively maintained. And the detailed information of the mountain recorded by the image is better restored, which can provide a real data basis for slope monitoring.

Nighttime infrared ship target detection based on Two-channel image separation combined with saliency mapping of local grayscale dynamic range

The automatic detection of infrared ship targets at night is the key research of intelligent maritime monitoring. Because the infrared maritime small targets are always dim, owning only simple texture, and variable in shape in voyage, deep learning is not so suitable for these kinds of target detection. In this paper, two-channel image separation combined with saliency mapping of local grayscale dynamic range (TCS-SMoLGDR) algorithm for ship target detection is proposed. The saliency mapping of local grayscale dynamic range (SMoLGDR) is proposed by taking full advantage of the feature that the infrared ship target at night owns the largest grayscale dynamic range. A saliency map is generated through SMoLGDR, in which the target area is significantly enhanced and the background is effectively suppressed. On the basis of the saliency map, the connected domain mean strategy is used to determine the real target area, which is beneficial to retain the complete target. For better segmentation of the target with uneven grayscale distribution from the background, this paper proposes a two-channel image separation (TCS) method to separate the local image of the target area into a bright channel image and a dark channel image, so that the grayscale distribution of targets in sub-images becomes relatively uniform. Finally, edge-guided binarization is used to extract the target. The experiment results show fully that the algorithm proposed in this paper can achieve the effective detection of the small dim targets in the nighttime infrared maritime image, and the detection accuracy is better than the comparison algorithms.

Nighttime Image Dehazing Based on Point Light Sources

Images routinely suffer from quality degradation in fog, mist, and other harsh weather conditions. Consequently, image dehazing is an essential and inevitable pre-processing step in computer vision tasks. Image quality enhancement for special scenes, especially nighttime image dehazing is extremely well studied for unmanned driving and nighttime surveillance, while the vast majority of dehazing algorithms in the past were only applicable to daytime conditions. After observing a large number of nighttime images, artificial light sources have replaced the position of the sun in daytime images and the impact of light sources on pixels varies with distance. This paper proposed a novel nighttime dehazing method using the light source influence matrix. The luminosity map can well express the photometric difference value of the picture light source. Then, the light source influence matrix is calculated to divide the image into near light source region and non-near light source region. Using the result of two regions, the two initial transmittances obtained by dark channel prior are fused by edge-preserving filtering. For the atmospheric light term, the initial atmospheric light value is corrected by the light source influence matrix. Finally, the final result is obtained by substituting the atmospheric light model. Theoretical analysis and comparative experiments verify the performance of the proposed image dehazing method. In terms of PSNR, SSIM, and UQI, this method improves 9.4%, 11.2%, and 3.3% over the existed night-time defogging method OSPF. In the future, we will explore the work from static picture dehazing to real-time video stream dehazing detection and will be used in detection on potential applications.