Small object detection has been a focus of attention since the emergence of deep learning-based object detection. Although classical object detection frameworks have made significant contributions to the development of object detection, there are still many issues to be resolved in detecting small objects due to the inherent complexity and diversity of real-world visual scenes. In particular, the YOLO (You Only Look Once) series of detection models, renowned for their real-time performance, have undergone numerous adaptations aimed at improving the detection of small targets. In this survey, we summarize the state-of-the-art YOLO-based small object detection methods. This review presents a systematic categorization of YOLO-based approaches for small-object detection, organized into four methodological avenues, namely attention-based feature enhancement, detection-head optimization, loss function, and multi-scale feature fusion strategies. We then examine the principal challenges addressed by each category. Finally, we analyze the performance of these methods on public benchmarks and, by comparing current approaches, identify limitations and outline directions for future research.

Cybersecurity situational awareness technology plays a critical role in assessing network security status, predicting potential attack paths, and assisting administrators in implementing effective defenses. Traditional methods for network situation assessment mostly rely on theoretical analysis, limiting their practicality in real-world networks. Additionally, the complexity of sensor-collected data often results in excessive storage demands. To address these challenges, this paper proposes a dynamic network attack-defense perception model that integrates reinforcement learning and game theory to enhance situational awareness and predict potential attack paths. The approach begins with the design of a hierarchical analytic process using a priority relation matrix to calculate system losses and assess security posture. Next, the Boltzmann probability distribution is employed to calculate the mixed-strategy Nash equilibrium, identifying optimal strategic responses. Finally, an improved Q-learning algorithm, in combination with game-theoretic principles, is used to dynamically model network state transitions, enabling accurate prediction of attack paths and supporting defenders in selecting optimal defense strategies. Simulation results validate the model’s effectiveness and practicality in complex network environments.