Research on SAR Image Ship Target Detection Algorithm with Improved YOLOv7

To address the issue of low detection accuracy in SAR image ship detection tasks, particularly in complex environments like small target ships and coastal ports, an improved YOLOv7-based algorithm is proposed. The bi Directional Feature Pyramid Network is introduced to simplify the network structure and improve the performance and feature representation ability of the network. The introduction of EIOU loss function is to replace the CIOU loss function used in the original model. This improvement makes the object detection network perform better in handling multi-scale targets, thereby reducing the problem of error detection and localization. Experimental results demonstrate that the improved model achieves a 1.08% increase in mAP on the SAR Chip Dataset, confirming the enhanced target detection performance of ships in SAR images.