Attribute-Guided Generative Adversarial Network With Improved Episode Training Strategy for Few-Shot SAR Image Generation

Abstract

Deep-learning-based models usually require a large amount of data for training, which guarantees the effectiveness of the trained model. Generative models are no exception, and sufficient training data are necessary for the diversity of generated images. However, for synthetic aperture radar (SAR) images, data acquisition is expensive. Therefore, SAR image generation under a few training samples is still a challenging problem to be solved. In this article, we propose an attribute-guided generative adversarial network (AGGAN) with an improved episode training strategy for few-shot SAR image generation. First, we design the AGGAN structure, and spectral normalization is used to stabilize the training in the few-shot situation. The attribute labels of AGGAN are designed to be the category and aspect angle labels, which are essential information for SAR images. Second, an improved episode training strategy is proposed according to the characteristics of the few-shot generative task, and it can improve the quality of generated images in the few-shot situation. In addition, we explore the effectiveness of the proposed method when using different auxiliary data for training and use the Moving and Stationary Target Acquisition and Recognition benchmark dataset and a simulated SAR dataset for verification. The experimental results show that AGGAN and the proposed improved episode training strategy can generate images of better quality when compared with some existing methods, which have been verified through visual observation, image similarity measures, and recognition experiments. When applying the generated images to the 5-shot SAR image recognition problem, the average recognition accuracy can be improved by at least 4$\%$.

Keywords

• Few-shot image generation
• generative adversarial network (GAN)
• meta-learning
• synthetic aperture radar (SAR)
• transfer learning