RDB-DINO: An Improved End-to-End Transformer With Refined De-Noising and Boxes for Small-Scale Ship Detection in SAR Images
Chuan Qin, Linping Zhang, Xueqian Wang, Gang Li, You He, Yuhui Liu
Abstract
Recently, convolution neural networks (CNNs) have been extensively utilized in synthetic aperture radar (SAR) ship detection owing to their strong feature extraction and representation capability. However, existing CNN-based SAR ship detectors often suffer from poor sensitivity to small-scale ship targets due to the limited extractable features, especially in complex inshore scenarios. Moreover, the hand-designed components like nonmaximum suppression (NMS) calculation and anchor generation imposed in CNN-based detector significantly affect their robustness. In the face of these challenges, a novel end-to-end (E2E) transformer-based detection framework for small-scale ship targets in SAR images, named detection transformer (DETR) with improved de-noising (DN) anchor box (DINO) with refined DN and box (RDB-DINO), is proposed in this article. First, we introduce a complete contrastive DN (CCD) training technique which reconstructs and exploits different kinds of noised queries to reduce the confusion between small ships and complex backgrounds. Second, a look twice toward maximum (LTTM) algorithm for iterative box refinement is designed to mine the abnormal sample information and obtain abundant features of small ships in the training process. Finally, substantial experiments conducted on two widely used open SAR ship datasets demonstrate that the proposed approach yields superior results in small ship detection performance, outperforming prevailing state-of-the-art (SOTA) benchmarks.