Bilinear constraint based ADMM for mixed Poisson-Gaussian noise removal

In this paper, we propose new operator-splitting algorithms for the total variation regularized infimal convolution (TV-IC) model [6] in order to remove mixed Poisson-Gaussian (MPG) noise. existing splitting algorithm TV-IC, an inner loop by Newton method had be adopted one nonlinear optimization subproblem, which increased computation cost per outer loop. By introducing a bilinear constraint and applying alternating direction of multipliers (ADMM), all subproblems proposed named as BCA (short Bilinear Constraint based ADMM algorithm) BCA$ _{f} $ variant with {\bf f} $ully form) can very efficiently solved. Especially $, they calculated without any iterations. The convergence are investigated, where particularly, Huber type TV regularizer is guarantee _f $. Numerically, compared primal-dual TV-IC model, algorithms, fewer tunable parameters, converge much faster produce comparable results meanwhile.