Aiming to solve the problems of false detection, missed detection, and insufficient detection ability of infrared vehicle images, an infrared vehicle target detection algorithm based on the improved YOLOv5 is proposed. The article analyzes the image characteristics of infrared vehicle detection, and then discusses the improved YOLOv5 algorithm in detail. The algorithm uses the DenseBlock module to increase the ability of shallow feature extraction. The Ghost convolution layer is used to replace the ordinary convolution layer, which increases the redundant feature graph based on linear calculation, improves the network feature extraction ability, and increases the amount of information from the original image. The detection accuracy of the whole network is enhanced by adding a channel attention mechanism and modifying loss function. Finally, the improved performance and comprehensive improved performance of each module are compared with common algorithms. Experimental results show that the detection accuracy of the DenseBlock and EIOU module added alone are improved by 2.5% and 3% compared with the original YOLOv5 algorithm, respectively, and the addition of the Ghost convolution module and SE module alone does not increase significantly. By using the EIOU module as the loss function, the three modules of DenseBlock, Ghost convolution and SE Layer are added to the YOLOv5 algorithm for comparative analysis, of which the combination of DenseBlock and Ghost convolution has the best effect. When adding three modules at the same time, the mAP fluctuation is smaller, which can reach 73.1%, which is 4.6% higher than the original YOLOv5 algorithm.

In view of the diverse jamming environment, the single‐level anti‐jamming method faces some challenges such as poor timeliness, high cost and poor effect. In this paper, routing delay, cost overhead and diversified jamming threats are comprehensively considered. In addition, a cross‐layer anti‐jamming method is proposed in the scenario of busy satellite Internet communication. The proposed cross‐layer anti‐jamming method involves two levels: link‐layer anti‐jamming based on path repair and network‐layer anti‐jamming based on path reconstruction. On the one hand, the channel is selected based on the improved Q‐learning anti‐jamming algorithm to confront common jamming. On the other hand, the route from the source to the destination node is selected based on the cross‐layer anti‐jamming algorithm to confront high‐intensity jamming. Finally, the simulation results show that, compared with other anti‐jamming algorithms, the proposed algorithm can achieve higher efficiency, lower cost, and more robust anti‐jamming routing.

In this paper, the dynamic spectrum access is investigated in satellite internet. Firstly, to describe the confrontation characteristics of the electromagnetic environment, two opposing teams denoted as the blue team (BT) and red team (RT), are designed. In addition, an “electromagnetic barrier” jamming strategy is proposed to achieve the effect of promoting approach defence with the attack. Precisely, the “electromagnetic barrier” can significantly weaken the communication ability of enemies (i.e. RT) while protecting the (i.e. BT) communication quality. Secondly, internal interference, external interference and malicious jamming of BT and RT are considered, and expected weighted aggregate interference and jamming (EWAIJ) are introduced to integrate these interference. Moreover, minimising interference is taken as the optimisation goal. Thirdly, Game theory is introduced to describe the combinatorial optimisation problem. The process of each team finding the optimal spectrum access strategy is proved to be an exact potential energy game. These two sub‐games together constitute the Stackelberg game framework. Finally, a distributed hierarchical confrontation channel selection algorithm (DHCCSA) is proposed to find the Stackelberg equilibrium solution. The simulation results show that the proposed algorithm can converge to a better effect compared with other algorithms. In addition, with the assistance of the “electromagnetic barrier,” BT is more robust than RT in terms of convergence rate, total utility, and total network throughput.

Camera failure or loss of storage components in imaging equipment may result in the loss of important image information or random pulse noise interference. The low-rank prior is one of the most important priors in image optimization processing. This paper reviews and compares some low-rank constraint models for image matrices. Firstly, an overview of image-inpainting models based on nuclear norm, truncated nuclear norm, weighted nuclear norm, and matrix-factorization-based F norm is presented, and corresponding optimization iterative algorithms are provided. Then, we use different image matrix low-order constraint models to recover satellite images from three types of pulse interference and provide our experimental visual and numerical results. Finally, it can be concluded that the method based on the weighted nuclear norm can achieve the best image restoration effect. The F norm method based on matrix factorization has the shortest computational time and can be used for large-scale low-rank matrix calculations. Compared with nuclear norm-based methods, weighted nuclear norm-based methods and truncated nuclear norm-based methods can significantly improve repair performance.

Camera malfunctions or loss of storage elements in imaging devices may lead to loss of important image information or random impulse noise interference. Low rank is one of the important prior information in image optimization processing. This paper uses different low-rank constraint models of the image matrices to recover the impulse interference satellite images. Firstly, an overview of image inpainting models based on nuclear norm, truncated nuclear norm, weighted nuclear norm, and matrix-factorization-based F-norm is presented, and corresponding optimiza-tion iterative algorithms are provided. Then, we conducted experiments under three types of pulse interference and provided visual and numerical results. Finally, it is concluded that the WSVT_ADMM method based on the weighted nuclear norm can obtain the best image inpainting results; The UV_ADMM method based on F norm of matrix factorization has the least computa-tion time and can be used for large-scale low-rank matrix computation; The WSVT_ADMM method based on weighted nuclear norm and the TSVT_ ADMM method based on truncated nuclear norms can significantly improve the repair effect compared to the nuclear norm-based methods such as SVT, SVP, and n_ADMM.