A Novel Multiscale Contrastive Learning Network for Fine-Grained Ocean Ship Classification

Fine-grained ocean ship classification plays a crucial role in maritime military surveillance, traffic management, and antismuggling operations. However, the complex backgrounds of remote sensing images (RSIs), as well as significant interclass similarities and intraclass differences, result in poor classification performance. Hence, we propose MSCL-Net, a multiscale contrastive learning network for fine-grained ship classification (FGSC). First, we introduce ResNet50 as the backbone network and extract the multilayer features by using the FPN for FGSC. Second, a channel spatial attention module (CSAM) is proposed to extract the similarity (contrastive) feature of the same class, strengthening the representation learning ability for addressing issues caused by significant interclass similarity and intraclass difference. Third, a region cropping and enlargement module is proposed to extract the fine-grained features of local discriminant regions in RSIs to overcome the challenge of background complexity. Finally, we used the CSAM to fuse the features of the original image and the cropped region image for FGSC. In addition, we introduce a combined loss based on center loss and PolyLoss to enhance the discrimination ability of features and make it more suitable for the imbalance dataset compared with cross-entropy. We use a public FGSC dataset, FGSC-23, and our FGSC-41 to evaluate the performance of MSCL-Net. The experimental results show superior performance compared to other state-of-the-art methods, highlighting the effectiveness of MSCL-Net in addressing the challenges associated with FGSC. Ablation experiments also suggest the effectiveness of our design.