CSPA-SCA: An Attention-Enhanced YOLOv11 Architecture for Small Ship Detection in Complex Maritime Scenarios

With the rapid development of maritime transportation, ship detection and recognition have become increasingly important for ocean management and national security. However, detecting ships in complex maritime environments poses significant challenges due to the small size of targets, low signal-to-noise ratio, and variations in appearance caused by changes in speed and position. In response to these challenges, this study leverages the YOLOv11 architecture and presents an improved model specifically designed for high-precision detection of small ship targets. We introduce the Cross-Stage Polarized Attention Spatial-Channel Attention (CSPA-SCA) module, which integrates cross-stage feature fusion and dual attention mechanisms to effectively enhance the model's multi-scale feature extraction and its ability to amplify blurred features. By incorporating this module into YOLOv11, the proposed algorithm significantly improves the accuracy and robustness of ship target detection in complex marine environments. In addition, we construct a new ship detection dataset that closely reflects real-world maritime conditions. Experimental results demonstrate that the proposed method achieves notable performance improvements on relevant datasets and exhibits strong potential for practical applications.