A robust variational model for positive image deconvolution

摘要: In this paper, an iterative method for robust deconvolution with positivity constraints is discussed. It based on the known variational interpretation of Richardson–Lucy as fixed-point iteration minimisation information divergence functional under a multiplicative perturbation model. The asymmetric penaliser function involved in then modified into and complemented regulariser. resulting gives rise to that we call regularised deconvolution. achieves image restoration quality comparable state-of-the-art computational efficiency similar original method. Experiments synthetic real-world data demonstrate performance proposed