Airborne radar forward-looking image enhancing algorithm based on generative adversarial networks

Abstract

Radar forward-looking imaging is gaining significance in various applications like battlefield reconnaissance, target surveillance, and precision guidance. Although synthetic aperture radar techniques provide high azimuth resolution but faced limitations in forward-looking area due to the poor Doppler resolution and the “left-right” ambiguity problem. Recently, generative adversarial networks have been extensively used for image motion blur removal. This letter proposes an end-to-end forward-looking image enhancing network using generative adversarial network to produce high-resolution images, improving the efficiency, and quality of imaging. Compared to conventional methods such as the deconvolution-based methods, this algorithm eliminates the need for design and iterative processes of the observation matrix. Simulated and real radar data validate that this approach offers robust recovery and better performance.