SYNTHETIC APERTURE RADAR IMAGE ENHANCEMENT BASED ON RESIDUAL NETWORK

Spatial resolution of synthetic aperture radar (SAR) is a vital index to evaluate the performance of its observed image. However, high spatial resolution of SAR is achieved at the cost of system resources. Therefore, super-resolution methods can be applied in SAR systems to improve their spatial resolution without system resource increases. In this paper, we propose a new residual network-based structure for super-resolution of SAR images. The proposed method adopts the structure of global residuals and adds several convolutional layers before and after the residual module to take into account the depth and width of the network. The simulation results show that the proposed method is effective as the visual effect and data evaluation.