Wang Jinyu, Yang Haitao, Liu Zhengjun, Chen Hang
Space Engineering University, Beijing, 101416, China.
School of Physics, Harbin Institute of Technology, Harbin, 150001, China.
Sci Rep. 2025 Mar 29;15(1):10867. doi: 10.1038/s41598-025-95106-7.
The limited availability of high-quality SAR images severely affects the accuracy and robustness of target detection, classification, and segmentation. To solve this problem, a novel image generation method based on a diffusion model is introduced that requires only one training sample to generate a realistic SAR image. We propose a single-scale architecture to avoid image noise accumulation. In addition, an attention module for the sampling layer in the generator for improving feature extraction is designed. Then, an information-guided attention module is proposed to suppress redundant information. Ship targets were selected as the research objects, and the proposed method was tested using an open-source dataset. We also built our own Sentinel-1 dataset to increase the number of challenges. The experimental results show that our method is optimal compared with the classical method SinGAN. Specifically, the SIFID is decreased from 4.80 × 10^(-4) to 1.66 × 10^(-7), the SSIM is improved from 0.07 to 0.51, and the LPIPS is decreased from 0.61 to 0.23. Compared with that of ExSinGAN, generation diversity increases by 27.35%.
高质量合成孔径雷达（SAR）图像的有限可用性严重影响目标检测、分类和分割的准确性和稳健性。为了解决这个问题，引入了一种基于扩散模型的新型图像生成方法，该方法仅需一个训练样本就能生成逼真的SAR图像。我们提出了一种单尺度架构以避免图像噪声积累。此外，设计了一种用于生成器中采样层的注意力模块，以改进特征提取。然后，提出了一种信息引导注意力模块来抑制冗余信息。选取舰船目标作为研究对象，并使用开源数据集对所提方法进行测试。我们还构建了自己的哨兵-1数据集以增加挑战性。实验结果表明，与经典方法SinGAN相比，我们的方法是最优的。具体而言，结构相似性指数偏差（SIFID）从4.80×10^(-4)降至1.66×10^(-7)，结构相似性（SSIM）从0.07提高到0.51， Learned Perceptual Image Patch Similarity（LPIPS）从0.61降至0.23。与ExSinGAN相比，生成多样性提高了27.35%。
