Multiframe resolution-enhancement using a robust iterative SRR based on leclerc stochastic technique

Abstract

This paper proposes a multiframe resolution-enhancement using a robust iterative SRR (Super-Resolution Reconstruction) for applying on images that is corrupted by several nose models. Typically, the success of SRR algorithms is highly dependent on the model accuracy regarding the imaging process. The real noise models corrupting the measure sequence are unknown hence SRR algorithms using L1 or L2 norm may degrade the image sequence rather than enhance it. The proposed enhancement algorithm is based on the stochastic regularization SRR technique of Bayesian MAP estimation by minimizing a cost function. The Leclerc norm is used for removing outliers in the data and for measuring the difference between the projected estimate of the HR image and each LR image. Due to the ill-pose problem, Tikhonov regularization is used to remove artifacts from the final answer and improve the rate of convergence. The experimental results show the effectiveness of our methods and demonstrate its superiority to other SRR algorithm based on L1 and L2 norm for several noise models such as Noiseless, AWGN, Poisson Noise, Salt&Pepper Noise and Speckle Noise.