ISAR High Resolution Imaging Based on Fast Sparse Bayesian Learning

Abstract

Aiming at the problems of poor ISAR imaging quality and long computation time with sparse apertures, a high resolution imaging algorithm based on fast sparse Bayesian learning is proposed. Firstly, an imaging model with sparse apertures based on compressed sensing is established. Then, it is assumed that each pixel of the target image obeys Gaussian prior to establish a sparse Bayesian model, and the fast marginal likelihood maximization method is used to reconstruct the high quality target image. Compared with the traditional sparse Bayesian learning method, the fast algorithm proposed can shorten the computation time while ensuring the quality of reconstruction. Finally, simulation experiments verify the effectiveness and superiority of the algorithm.