Video Enhancement Using a Robust Iterative SRR Based on a Geman&McClure Stochastic Estimation

Most video enhancement algorithms assume that the noise model of the imaging system is known as AWGN thereby imaging process model violations often occur since the real noise model is not known in many practical applications. Robust statistics has emerged as a family of theories and techniques for estimation while dealing with deviations from the idealized model assumptions. In this paper, we propose a novel robust video enhancement algorithm using SRR (Super-Resolution Reconstruction) based on the stochastic regularization technique by minimizing a cost function. First, the registration process is used to estimate the relationship between the reference frame and other neighboring frames. Then, the Geman&McClure norm is used for measuring the difference between the projected estimate of the high quality image and each low high quality image and for removing outliers in the data. Moreover, Tikhonov regularization is incorporated in the proposed framework in order to remove artifacts from the final answer and to improve the rate of convergence. Finally, experimental results are presented to demonstrate the outstanding performance of the proposed algorithm in comparison to several previously published methods using standard sequences such as Foreman and Susie that are corrupted by several noise models such as AWGN, Poisson Noise, Salt & Pepper Noise and Speckle Noise.