A Coordinate Attention-Enhanced YOLOv8 Method for Detecting Damage States of Ship Mooring Ropes

The integrity of ship mooring ropes is critical to the safe operation of waterborne autonomous transportation systems. To address the challenges of low detection accuracy and poor anti-interference capability in complex and variable environments, this paper proposes an innovative CA-YOLOv8 detection method. Based on the YOLOv8 deep learning network, the method constructs a detection model covering four damage states: complete, mild abrasion, severe abrasion, and broken. By incorporating the Coordinate Attention (CA) mechanism, the model can precisely focus on subtle damage regions of mooring ropes, effectively suppress interference from complex backgrounds such as waves and lighting, and enhance the extraction of spatial structure and channel context information, thereby improving the efficiency of damage state detection. Experimental results demonstrate that the proposed method achieves outstanding performance in detecting multiple damage states of ship mooring ropes under complex backgrounds and challenging scenarios, with recall and mAP reaching 88.1% and 91.2%, respectively, improving the baseline model by 7.2% and 1.5%. This study provides efficient and reliable technical support for the health maintenance of mooring ropes and the safety guarantee of ship navigation, which holds considerable implications for advancing the intelligence of waterway transportation systems.