Performance Study of the Robust Bayesian Regularization Technique for Remote Sensing Imaging in Geophysical Applications

Abstract

In this paper, a performance study of a methodology for reconstruction of high-resolution remote sensing imagery is presented. This method is the robust version of the Bayesian regularization (BR) technique, which performs the image reconstruction as a solution of the ill-conditioned inverse spatial spectrum pattern (SSP) estimation problem with model uncertainties via unifying the Bayesian minimum risk (BMR) estimation strategy with the maximum entropy (ME) randomized a priori image model and other projection-type regularization constraints imposed on the solution. The results of extended comparative simulation study of a family of image formation/enhancement algorithms that employ the RBR method for high-resolution reconstruction of the SSP is presented. Moreover, the computational complexity of different methods are analyzed and reported together with the scene imaging protocols. The advantages of the remote sensing imaging experiment (that employ the BR-based estimator) over the cases of poorer designed experiments (that employ the conventional matched spatial filtering as well as the least squares techniques) are verified trough the simulation study. Finally, the application of this estimator in geophysical applications of remote sensing imagery is described.