SAR Image Synthesis with Diffusion Models

TL;DR

This paper adapts denoising diffusion probabilistic models to synthetic aperture radar (SAR) image generation, demonstrating superior quality over GANs and benefits from pretraining on large clutter datasets.

Contribution

It introduces the application of DDPM to SAR imaging, optimizing network and diffusion parameters, and shows improved results over existing GAN-based methods.

Findings

DDPM outperforms GANs in SAR image quality

Pretraining on large clutter data enhances image realism

Qualitative and quantitative improvements over state-of-the-art methods

Abstract

In recent years, diffusion models (DMs) have become a popular method for generating synthetic data. By achieving samples of higher quality, they quickly became superior to generative adversarial networks (GANs) and the current state-of-the-art method in generative modeling. However, their potential has not yet been exploited in radar, where the lack of available training data is a long-standing problem. In this work, a specific type of DMs, namely denoising diffusion probabilistic model (DDPM) is adapted to the SAR domain. We investigate the network choice and specific diffusion parameters for conditional and unconditional SAR image generation. In our experiments, we show that DDPM qualitatively and quantitatively outperforms state-of-the-art GAN-based methods for SAR image generation. Finally, we show that DDPM profits from pretraining on largescale clutter data, generating SAR images of even higher quality.