An efficient projected subgradient algorithm for blind image deconvolution using an L1-TV cost function
Wenyao Xia, Dimitrios Hatzinakos · 2012
Traditional blind image iterative algorithms are designed for Gaussian noise by using the L2-norm error term. For robustness against the influence of non-Gaussian noise, an efficient projected subgradient algorithm for blind image deconvolution is developed, based on a TV cost function with the L1-norm error term. Because of using the subgradient technique, the proposed subgradient algorithm can minimize the L1-TV cost function directly. By contrast, existing L1norm-based image restoration algorithms only minimize the approximate L1cost function and assume a known blur. Illustrative examples show that under suboptimal regularization parameters, the projected subgradient algorithm is efficient in producing better image estimate than two traditional blind image iterative algorithms in terms of both ISNR and perception.