| Popis: |
In this paper, we propose a parameter-free image reconstruction algorithm that is based on the maximum a posteriori (MAP) estimation method and so called "integrate-out" approach for determining the prior distribution. The resulting objective function is minimized using the majorize-minimize optimization technique. The proposed algorithm avoids the computational burden associated with using cross-validation methods for choosing the penalty parameter in the popular least absolute shrinkage and selector operator (LASSO) method. We tested the algorithm on synthetic data obtained from the U.S Army Research Laboratory (ARL) in Adelphi, MD. As compared to the image reconstructed using the delay-and-sum algorithm, the reconstructed images is more sparse, with suppressed side lobes and background noise. |
