Multi-angle Image Fusion in Ground-based Radar Lunar Imaging Based on Image Kurtosis

The fusion of multi-angle lunar images can provide more comprehensive lunar topography information than original images, and ground-based (GB) radar has the ability to acquire multi-angle lunar images. In GB radar multi-angle lunar image fusion, proper radar trajectories should be selected, and image fusion algorithm should be used. In this paper, the multi-angle lunar image fusion technique for GB synthetic aperture radar (SAR) based on image kurtosis is proposed. To begin with, the echo signal model considering space-variant backscattering coefficient is established to reflect lunar topography feature. After that, a multi-angle radar trajectory selection method based on K-means algorithm is proposed to acquire multi-angle lunar images. Thereafter a multi-angle lunar image fusion algorithm based on image kurtosis and gradient descent is proposed to acquire the fused image. Finally, computer simulation results validate the effectiveness of the proposed technique.