Unmanned aerial vehicle swarm （UAVS） operation is a new operation mode in the future war. The resilience reconstruction capability of the UAVS network is a key index to evaluate the combat effectiveness of the UAVS system. The resilience reconstruction mechanism of the UAVS network is analyzed. A community network model of UAVS is designed to improve network connectivity. Since the network performance of UAVS is sharply degraded under malicious attack， the minimum threshold of information network interaction capability is proposed， and the resilience evaluation method of swarm network performance at the failure stage is optimized. Finally， the average shortest path， information interaction capability， network clustering coefficient， and other parameters of the network under different scales are compared by simulation modeling. The swarm network is reconstructed under the condition of multi-type and multi-scale damage， and the impact on network resilience is evaluated. The experimental results show that the proposed model has higher resilience. The research results can improve the ability of UAVS to cope with complex disturbances and provide theoretical guidance for the multi-task planning of UAVS.

The intelligent application of unmanned aerial vehicles （UAVs） is a hot topic in current research. To achieve efficient and real-time air-to-ground target detection for UAVs， this paper proposes a lightweight MBCF-YOLO model for air-to-ground target detection. First， a deep separable convolution to redesign an efficient and lightweight backbone network is introduced. Then， a small target detection branch with embedded recursive attention mechanism is added to the neck network to improve the detection accuracy of small targets and the ability of feature fusion. Finally， the focus loss function is combined with loss rank mining to improve the original loss function to overcome the data imbalance problem of air-to-ground datasets. Several control experiments and real machine applications have been conducted on this algorithm model， and the results show that compared with other current algorithms， the MBCF-YOLO algorithm achieves a better balance between accuracy and latency on the VisDrone 2021 dataset.