ISAR Resolution Enhancement Method Exploiting Generative Adversarial Network

Deep learning has been used in inverse synthetic aperture radar (ISAR) imaging to improve resolution performance, but there still exist some problems: the loss of weak scattering points, over-smoothed results, and universality generalization. To address these problems, an ISAR enhancement method exploiting a generative adversarial network (GAN) is proposed this paper. We adopt relativistic average discriminator (RaD) enhance ability describe target details. The function composed feature loss, absolute loss. get main characteristics target. ensures that GAN recovers more adopted make results not over-smoothed. Experiments based on simulated measured data under different conditions demonstrate good performance. In addition, generalization are also well verified.