A review of YOLOv11 based on sar ship detection

Abstract

Synthetic Aperture Radar (SAR) imaging technology holds significant value in military reconnaissance and civil maritime monitoring due to its all-weather and all-time imaging capabilities. Ship detection, as a core task of maritime monitoring, plays a crucial role in ensuring maritime safety, combating illegal fishing, protecting the marine environment, and military target reconnaissance. However, inherent issues in SAR images, such as noise, the difficulty of detecting small targets, and interference from complex sea conditions, pose significant challenges to the design of ship detection algorithms. In recent years, the YOLO series of algorithms has continuously evolved in SAR ship detection, with the latest version, YOLOv11, effectively enhancing detection accuracy and efficiency through innovations such as lightweight design, multi-scale feature modeling, and improved attention mechanisms. This paper starts with the key technologies in SAR ship detection, providing a detailed analysis of the performance advantages of YOLOv11 and its comparison with YOLOv10. The results indicate that YOLOv11 significantly outperforms YOLOv10 in core metrics such as precision, recall, and mAP50, while also exhibiting faster convergence and stronger generalization capabilities. Additionally, this paper explores the application prospects of YOLOv11 in complex scenarios and proposes future optimization directions regarding the physical modeling and adaptability issues of SAR images, providing important references for solving target detection problems in complex environments.