Traditional methods of generating digital camouflages cannot generate digital camouflages based on the background information in real time. In order to cope with this problem， a digital camouflage generation method based on cycle-consistent adversarial network was proposed. Firstly， the image features were extracted by using densely connected convolutional network， and the learned digital camouflage features were mapped into the background image. Secondly， the color retention loss was added to improve the quality of generated digital camouflages， ensuring that the generated digital camouflages were consistent with the surrounding background colors. Finally， a self-normalized neural network was added to the discriminator to improve the robustness of the model against noise. For the lack of objective evaluation criteria for digital camouflages， the edge detection algorithm and the Structural SIMilarity （SSIM） algorithm were used to evaluate the camouflage effects of the generated digital camouflages. Experimental results show that the SSIM score of the digital camouflage generated by the proposed method on the self-made datasets is reduced by more than 30% compared with the existing algorithms， verifying the effectiveness of the proposed method in the digital camouflage generation task.

In view of the problem that the traditional multi-Agent flocking algorithms are not universal when a single target tracking is considered, and the existing multi-target flocking control is controlled by centralized coordinated movement based on global target information, rather than the distributed coordinated control based on local destination information. Therefore, a distributed motion cooperative pinning flocking algorithm combined with local adaptive tracking was presented. First, the local adaptive tracking strategy based on separation, aggregation, velocity matching and direct feedback was introduced to achieve local following interaction dynamically. Secondly, a node influence index evaluation algorithm based on pinning idea was presented to select the m information Agents to track m targets, playing an important role in simulating external information; different information individuals indirectly lead individuals with a different target to track the respective target due to local adaptive detection mechanism. Finally, a new class of potential functions of aggregation and exclusion with the advantages of less adjustable parameters and high efficiency was designed; the Agents with same target could gather in the process of tracking, and the Agents with different target could avoid collision based on the potential function. The experimental results under three dimensional space show the feasibility and effectiveness of multi-target tracking.