Ship Target Detection Method Based on Infrared Wake Enhancement and Multimodal Fusion

To address the challenges of ship target detection in complex water environments, this paper proposes an improved YOLOv11-based ship detection method that fuses infrared and optical images. First, texture enhancement is performed on ship wake features in infrared images, leveraging their sensitivity to temperature differences and wake information under extreme conditions such as low visibility and backlighting. Then, the enhanced infrared images and optical images are jointly fed into a dual-branch feature extraction network. The fusion attention module adaptively highlights key regions and channels to achieve feature complementarity and deep fusion. Experimental results demonstrate that the proposed method significantly improves the accuracy and robustness of ship detection in various complex scenarios, fully validating the effectiveness of multisource fusion and wake feature enhancement strategies.