Image dehazing is essential to boost the visual quality of images captured in hazy conditions. Recently, many learning-based methods were proposed to achieve single image dehazing with the training of tremendous paired synthetic hazy/ real clean images. Due to the domain gap between real and synthetic scenes, these models cannot generalize well to various real hazy scenes, leading to under-dehazed results. To overcome this problem, we propose a real scene image Dehazing Network with Multi-prior Guidance and Domain Transfer (DNMGDT). Our DNMGDT is based on a parameter shared architecture trained by synthetic hazy images and real hazy images simultaneously. For real hazy images, multiple prior-based dehazed images are adopted as pseudo clean images. An Image Quality Guided Adaptive Weighting (IQGAW) scheme is proposed to form the supervision by automatically weighting different parts of these prior-based dehazed images and suppressing negative information of them. Moreover, to reduce the domain gap between real and synthetic hazy scenes, a Physical Model Guided image level Domain Transfer (PMGDT) mechanism is proposed to regularize the learning process with consistency constraint. Experiments on various datasets demonstrated the effectiveness of our proposed method especially for real hazy scenes.

As jamming systems continue to evolve, jammers are becoming increasingly complex. They modify their strategies in response to changes in radar status, thereby creating a dynamic confrontation with the radar. Currently, deep reinforcement learning (DRL) methods are widely employed in the decision-making process for radar anti-jamming strategies; however, most of these methods overlook the learning capabilities of the jammer. This letter proposes a novel multi-agent deep reinforcement learning (MADRL) method that leverages the robust capabilities of the neural fictitious self-play (NFSP) method and extensive-form game (EFG) frameworks to address multi-agent interactions and partial information observability challenges. The letter primarily discusses the model design, which includes the construction of the EFG framework, hierarchical optimization of the action space, and the design of the reward function. Comparative results with other state-of-the-art methods demonstrate that this approach achieves significant accuracy and rapid convergence in anti-jamming decision-making, thereby effectively enhancing the intelligent anti-jamming capabilities of radar systems.

Deep learning has become a promising tool for processing the massive data and attracted an increasing attention in the fields of degradation modeling and remaining useful life (RUL) prediction. The existing deep-learning-based methods are generally faced with the two aspects of problems. On the one hand, the prediction results are represented by the point estimates instead of the probabilistic distribution and, thus, the prognostic uncertainty in RUL prediction cannot be characterized. On the other hand, there exist plenty of the engineering assets without the label data related to lifetime, posing a great challenge for training the deep learning network. Toward this end, we propose a prognostic model under the framework of Bayesian deep learning for equipment lacking the label data related to lifetime. First, the monitoring data of the historical equipment and the historical data of field equipment in the database are preprocessed to generate the samples regarding degradation information as a label. Second, the bidirectional recurrent neural network (RNN) is employed as the candidate network for the advantages in handling the sequential monitoring data. On the basis of this, the idea of Bayesian deep learning is incorporated into the bidirectional RNN; thus, we can characterize the uncertainty of the predicted degradation level at any future time via utilizing the variational inference technique in the Bayesian neural networks. Furthermore, the failure probability for the concerned equipment at any time can be determined, by which the degradation uncertainty can be converted into the RUL uncertainty from the point of the reliability theory. Finally, we provide the case study associated with lithium-ion batteries to verify the proposed prognostic model.

End-to-end learning-based image dehazing methods tend to overdehaze or underdehaze in real scenes due to inefficient feature extraction and feature fusion. In this letter, we propose a multiscale supervision-guided context aggregation network (MSGCAN) based on two principles: improving feature extraction and enhancing feature mapping. To improve feature extraction, an attention-guided context aggregation (AGCA) module is adopted to merge context features extracted by several residual dense blocks (RDB). Moreover, we output these aggregated context features on each scale and form multiscale supervision to enhance feature mapping and ensure that the extracted features on each scale contain more realistic details. The experimental results show that the proposed MSGCAN performs better than other state-of-the-art dehazing methods in both synthetic and real-world scenes.

Abstract Single‐image dehazing is an important problem because it is a key prerequisite for most high‐level computer vision tasks. Traditional prior‐based methods adopt priors generated from clear images to restrain the atmospheric scattering model and then recover haze‐free images. However, these prior‐based methods always encounter over‐enhancement, such as halos and colour distortion. To solve this problem, many works use a convolutional neural network to retrieve original images. However, without priors as guidance, these learning‐based methods dehaze effectively in synthetic datasets but perform poorly in real scenes. Hence, in this paper, we propose a prior‐guided multiscale network for single‐image dehazing named PGMNet. Specifically, prior‐based methods are adopted to acquire dehazed images of the training dataset in advance and then send these dehazed images to a parameter‐shared encoder to form multiscale features. During the decoding process, these multiscale features are adopted to guide the prior‐guided multiscale network to recover more image details. Moreover, considering that these prior‐based dehazed images usually contain some over‐enhanced regions, a spatial attention guided feature aggregation module and squeeze‐and‐excitation module are adopted to alleviate colour distortion. The proposed PGMNet takes the advantage of prior‐based methods in real haze removal and provides superior performance compared with the state‐of‐the‐art methods on both synthetic and real‐world datasets.

Abstract The optimal design of transmitted waveforms to improve the performance of missile‐borne radar in suppressing interrupted sampling repeater jamming (ISRJ) is studied here. The waveform diversity property requires that the waveform group has excellent orthogonality (good auto‐ and cross‐correlation properties), which is an effective waveform optimisation design method for missile‐borne radar electronic counter‐countermeasures. The principle of orthogonal phase coding waveform against ISRJ is studied and the use of chaotic sequences is proposed to initialise the population, as well as a simulated annealing selection operator, which improves the convergence speed of the improved genetic algorithm (IGA) and enhances the global optimisation ability. The IGA improves the shortcomings of GA, that is, easily falling into a local optimum and slow search speed, and can realise a global search for longer orthogonal polyphase coding sequence sets. The simulation examples demonstrate the good jamming suppression performance of the proposed method.

A fast algorithm is proposed to tackle the constrained total generalised variation (TGV)‐based image‐restoration and reconstruction problems. The proposed algorithm proceeds by splitting: the non‐smooth constrained TGV model is first decomposed into several sub‐problems easier to solve, and then the linear gradient or proximity operators, including projections and shrinkages, of the sub‐problems are individually called without inner iteration. The algorithm is highly parallel since most of its steps can be executed simultaneously. Image‐restoration and reconstruction experiments demonstrate that the proposed algorithm outperforms several state‐of‐the‐art TV‐based methods both in accuracy and high‐speed efficiency. Besides, the proposed method efficiently suppresses staircase effects and presents better visual impression.

To solve the homography estimation problem containing outliers and noise, a fast, robust, and accurate method is proposed. In this method, the outliers are rejected based on the differing characteristics of algebraic errors between outliers and inliers, and the homography is estimated by minimising the residual vector. The advantage of this method is in integrating the outlier rejection into the estimation pipeline. The computational complexity of the proposed method is not increased, and the random sample consensus algorithm is not needed to extract the inliers, as was previously necessary. Since the outlier rejection process is based on an algebraic criterion without computing the re‐projection error at each step, the speed of the proposed method is improved. Several simulations based on synthetic and real images illustrate the performance of the proposed method in terms of subjective visual quality, objective quality measurement, and computational time. The experimental results demonstrate that the proposed method achieves accurate, efficient and robust homography estimation under different image transformation degrees and different outlier ratios.

A parallel linearized alternating direction method of multipliers (PLADMM) is proposed to solve large-scale imaging inverse problems, which involve the sum of several linear-operator-coupled nonsmooth terms. In the proposed method, the proximity operators of the nonsmooth terms are called individually at each iteration and the auxiliary variables existing in the classical ADMM are excluded. Therefore, the proposed method possesses a highly parallel structure and most of its substeps can be executed simultaneously. The application to multichannel total generalized variation (TGV) based image restoration shows the effectiveness of the proposed method.

Unlike the traditional wavelets, which lack the ability to detect directionality, shearlets provide a multidirectional as well as multiscale decomposition for multidimension signals. In this paper, we propose a new adaptive shearlet-regularized iterative image deblurring algorithm, where the image decon-volution problem is solved through the alternating direction method of multipliers (ADMM). As the discrepancy principle is employed, the regularization parameter can be refreshed with a closed form adaptively in each iteration. Numerical simulation results of image deblurring problems are presented to illustrate the effectiveness of the proposed method in terms of both detail preservation and speed.

Estimation of the regularization parameter, which strikes a balance between the data fidelity and regularity, is essential for successfully solving ill-posed image restoration problems. Based on the classical total variation (TV) model and prevalent alternating direction method of multipliers, we hammer out a fast algorithm being able to simultaneously estimate the regularization parameter and restore the degraded image. By applying variable splitting technique to both the regularization term and data fidelity term, we overcome the nondifferentiability of TV and achieve a closed form to update the regularization parameter in each iteration. The solution is guaranteed to satisfy Morozov's discrepancy principle. Furthermore, we present a convergence proof for the proposed algorithm on the premise of a variable regularization parameter. Experimental results demonstrate that the proposed algorithm is superior in speed and competitive in accuracy compared with several state-of-the-art methods. Besides, the proposed method can be smoothly extended to the multichannel image restoration.