Inverse synthetic aperture radar imaging method based on generative adversarial network

Abstract

The invention discloses an inverse synthetic aperture radar imaging method based on a generative adversarial network. A generative adversarial network (GAN) is composed of a generator network and a discriminator network. The generator network uses a convolution layer and a residual network module to extract feature representation and maintain low-dimensional feature information, and uses a deconvolution layer to reconstruct an ISAR target image. And the discriminator network extracts feature information from an ISAR image output by the generator network by using the convolutional layer, thereby realizing authenticity discrimination of the ISAR image. And at the network training stage, all layers of parameters of the generator network and the discriminator network are updated by using the training error output by the discriminator network. And the trained generator network is separated from the GAN and is used for under-sampling ISAR data imaging. At the imaging stage, a low-quality target image obtained by under-sampling ISAR target echo data through a distance-Doppler RD method is input into the generator network, and correspondingly, a high-quality ISAR target image is output. The imaging quality and the calculation efficiency of the method are superior to those of a traditional distance Doppler imaging method and a compressed sensing imaging result.

Description

technical field [0001] The invention relates to the technical field of radar signal processing, in particular to an inverse synthetic aperture radar (Inverse Synthetic Aperture Radar, ISAR) imaging method based on a generative adversarial network (Generative Adversarial Network, GAN). Background technique [0002] Inverse synthetic aperture radar is a typical imaging radar system, which is mainly used to obtain high-resolution images of non-cooperative moving targets, and is an effective means of target recognition. The traditional radar imaging method is the range-Doppler (RangeDoppler, RD) type imaging method, which uses the Doppler modulation echo signal within the coherent processing time (Coherent Processing Interval, CPI) to obtain high resolution in azimuth. [0003] In 2007, Professor Baraniuk and others introduced the (Compressive Sensing, CS) theory into the field of radar imaging. Since then, the CS-based ISAR imaging method has attracted more and more attention ...

AI Technical Summary

Problems solved by technology

But at the same time, the performance of the CS ISAR imaging method is still limited by the inaccurate sparse representation and the low efficiency of the reconstruction method.

Images