MSCF-Net: A Lightweight Multi-Scale SAR Image Oriented Ship Detection Algorithm in Complex Scenes

To address the challenges in synthetic aperture radar (SAR) ship detection, including complex target backgrounds, multiscale ships, and diverse orientations, while also meeting the lightweight requirements for satellite-based applications, we propose a lightweight multiscale feature fusion network—MSCF-Net. First, we design a lightweight dilation-wise residual C2f (DRC2F) module that enhances the network’s multiscale feature extraction capability through a two-stage residual mechanism and multiscale depth-wise separable dilated convolutions. Second, to overcome the limitations of traditional feature fusion methods in SAR ship detection, we propose a multiscale channel feature fusion pyramid network, MSCF-FPN, which effectively suppresses background noise interference while highlighting foreground target features through multimodal pooling and dynamic feature calibration mechanisms. Finally, to further improve detection accuracy on SAR images characterized by low target-background discriminability and diverse target orientations, we propose a multibranch decoupled detection head integrated with receptive-field attention to improve the detection head’s capacity to perceive spatial as well as orientation information. Experimental results demonstrate that MSCF-Net achieves detection accuracies of 79.1% (+4.9%) and 92.1% (+2%) on the SRSDD-v1.0 and high-resolution SAR images dataset (HRSID), respectively, with mean average precisions of 71.7% (+5.8%) and 93.3% (+0.2%). Furthermore, the number of model parameters is decreased from 10.89M to 4.66M, a reduction of approximately 57.2%, striking an effective balance between detection performance and model efficiency. In addition, MSCF-Net exhibits robust generalization capabilities on large-scene SAR images, rendering it well-suited for application in complicated real-world scenarios.