Self-Supervised-ISAR-Net Enables Fast Sparse ISAR Imaging

TL;DR

This paper introduces SS-ISAR-Net, a self-supervised neural network for sparse ISAR imaging that leverages the equivariant property to enable high-quality image reconstruction without paired training data.

Contribution

The paper proposes a novel self-supervised unfolded neural network, SS-ISAR-Net, which utilizes the equivariant property for sparse ISAR imaging, reducing the need for paired images during training.

Findings

Outperforms existing methods on synthetic data

Effective with real-world measurement data

Mitigates grating lobes in sparse ISAR imaging

Abstract

Numerous sparse inverse synthetic aperture radar (ISAR) imaging methods based on unfolded neural networks have been developed for high-quality image reconstruction with sparse measurements. However, their training typically requires paired ISAR images and echoes, which are often difficult to obtain. Meanwhile, one property can be observed that for a certain sparse measurement configuration of ISAR, when a target is rotated around its center of mass, only the image of the target undergoes the corresponding rotation after ISAR imaging, while the grating lobes do not follow this rotation and are solely determined by the sparse-sampling pattern. This property is mathematically termed as the equivariant property. Taking advantage of this property, an unfolded neural network for sparse ISAR imaging with self-supervised learning, named SS-ISAR-Net is proposed. It effectively mitigates grating lobes caused by sparse radar echo, allowing high-quality training to be achieved using only sparse radar echo data. The superiority of the proposed SS-ISAR-Net, compared to existing methods, is verified through experiments with both synthetic and real-world measurement data.