J Shanghai Jiaotong Univ Sci ›› 2024, Vol. 29 ›› Issue (4): 702-711.doi: 10.1007/s12204-024-2746-8

• Special Issue on Multi-Agent Collaborative Perception and Control • Previous Articles    

Data Augmentation of Ship Wakes in SAR Images Based on Improved CycleGAN

基于改进CycleGAN的SAR图像舰船尾迹数据增强

YAN Congqiang1,2 (鄢丛强), GUO Zhengyun3,4 (郭正玉), CAI Yunze1,2∗∗ (蔡云泽)   

  1. (1. Department of Automation, Shanghai Jiao Tong University, Shanghai 200240, China; 2. Key Laboratory of System Control and Information Processing of Ministry of Education, Shanghai 200240, China; 3. China Airborne Missile Academy, Luoyang 471009, Henan, China; 4. National Key Laboratory of Air-Based Information Perception and Fusion, Luoyang 471000, Henan, China)
  2. (1. 上海交通大学自动化系,上海200240;2. 系统控制与信息处理教育部重点实验室,上海200240;3. 中国空空导弹研究院,河南 洛阳 471009;4. 空基信息感知与融合全国重点实验室,河南 洛阳471000)
  • Accepted:2023-09-26 Online:2024-07-28 Published:2024-07-28

Abstract: The study on ship wakes of synthetic aperture radar (SAR) images holds great importance in detecting ship targets in the ocean. In this study, we focus on the issues of low quantity and insufficient diversity in ship wakes of SAR images, and propose a method of data augmentation of ship wakes in SAR images based on the improved cycle-consistent generative adversarial network (CycleGAN). The improvement measures mainly include two aspects: First, to enhance the quality of the generated images and guarantee a stable training process of the model, the least-squares loss is employed as the adversarial loss function; Second, the decoder of the generator is augmented with the convolutional block attention module (CBAM) to address the issue of missing details in the generated ship wakes of SAR images at the microscopic level. The experiment findings indicate that the improved CycleGAN model generates clearer ship wakes of SAR images, and outperforms the traditional CycleGAN models in both subjective and objective aspects.

Key words: synthetic aperture radar (SAR), ship wake, data augmentation, cycle-consistent generative adversarial network (CycleGAN), attention mechanism

摘要: 合成孔径雷达(SAR)图像的舰船尾迹研究在海洋船舶目标的探测中具有重要意义。本研究针对SAR图像舰船尾迹数据样本数量少和多样性不足的问题,提出一种基于改进的循环一致性生成对抗网络(CycleGAN)的SAR图像舰船尾迹数据增强方法。改进措施主要包括两个方面:第一,采用最小二乘损失作为对抗损失函数,提高了生成图像的质量,稳定了模型的训练过程;第二,在生成器的解码器中嵌入卷积块注意力模块(CBAM),在微观层面上解决了生成的SAR图像舰船尾迹中信息丢失的问题。实验结果表明,改进后的CycleGAN模型能生成更清晰的SAR图像舰船尾迹样本,在主观和客观方面都优于传统的CycleGAN模型。

关键词: 合成孔径雷达(SAR),舰船尾迹,数据增强,循环一致性生成对抗网络(CycleGAN),注意力机制

CLC Number: