LSDFormer: Lightweight SAR Ship Detection Enhanced With Efficient Multiattention and Structural Reparameterization

Ship detection in Synthetic Aperture Radar (SAR) images faces challenges such as strong background interference, varying ship appearance and distribution and high real-time requirements. Although attention-based deep learning methods dominate this field, the design of lightweight models with efficient attention mechanisms capable of addressing the above challenges remains underexplored. To address this issue, we propose a lightweight SAR ship detection model named LSDFormer, which is built upon the MetaFormer architecture and consists of an efficient multi-attention enhanced backbone and neck and a structural reparameterization enhanced head. We employ two lightweight modules for the backbone and neck: a PoolFormer-based feature extraction module with efficient channel modulation attention is proposed to enhance ship features and suppress background interference; a downsampling module using efficient channel aggregation attention and group convolutions is introduced to enrich ship features. The position-sensitive attention from YOLOv11 is also introduced to handle variations in ship appearance and distribution. These three attentions are integrated into an efficient multi-attention mechanism. Furthermore, a structural reparameterization based detection branch is proposed for the head of LSDFormer, which enhances ship features while reducing model complexity. Extensive experiments on SSDD and HRSID datasets demonstrate the superiority and effectiveness of LSDFormer, achieving AP50 of$\bf {98.5\pm 0.4\%}$and$\bf {92.8\pm 0.2\%}$, respectively, with only$\bf {1.5}$M parameters and$\bf {4.1}$GFLOPs. The average processing time per image is$\bf {4.9}$ms on SSDD and$\bf {4.2}$ms on HRSID, confirming its real-time performance.