Radar correlated imaging (RCI) is a novel modality to obtain high resolution target images by correlated process of stochastic radiation field and the received signals. Conventional RCI methods neglect the inherent structure information of complex extended target, which makes the quality of recovery result degraded. Thus a clustered sparse Bayesian learning with Laplace prior (La-CSBL) algorithm for extended target imaging is proposed in this paper. A hierarchical correlated Laplace prior model is introduced to consider both the sparse prior and the cluster prior, and the prior for each coefficient not only involves its own hyperparameter, but also its immediate neighbor hyperparameters. Then the algorithm alternates between steps of target reconstruction and parameter optimization by cyclic minimization method under the Bayesian maximum a posteriori framework. Experimental results show that the proposed algorithm could realize high resolution imaging efficiently for extended target.
Based on the temporal-spatial stochastic radiation field, microwave staring correlated imaging (MSCI) can achieve high resolution reconstruction of the target. When MCSI is applied in remote sensing, the imaging area will be very large. With the increasing of imaging equation scale, the ill-conditioned problem will seriously influence the resolution of imaging. This paper focuses on the algorithm to realize high-resolution MSCI in condition of large imaging area. Inspired by multigrid (MG) method, we use multigrid (MG) method to decrease the low-frequency error and highfrequency error of the solution by dealing with the imaging equation in different scales. Furthermore, we apply Least Square QR-factorization (LSQR) algorithm to solve the large scale ill-conditioned equation. We name the new algorithm as MG-LSQR. Simulation results show that the proposed algorithm could increase the resolution efficiently in condition of large imaging area.
Based on the temporal-spatial stochastic radiation field (TSSRF), microwave staring correlated imaging (MSCI) can achieve high resolution images of the targets. This paper focuses on the nonlocal and local regularization to improve the MSCI reconstruction performance. Low-rank regularization is considered to reveal the global information of the images. For local, total variation is a typical choice for its excellent edge preserving ability and noise reduction. Therefore, a method with the combination of the low-rank and total variation regularization is considered in this paper, in which a logarithmic function is taken as a non-convex approximation of the rank function. By elaborately adjusting the regularization parameters, the whole problem is still convex. Thus the Split Bregman method is utilized to solve the convex optimization problem and form the proposed algorithm, namely the LogRankTV algorithm. The effectiveness of the LogRankTV algorithm is demonstrated via the simulation results.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.