An improved model for restoring and protecting the digital images
Nour‐Eddine Joudar, Mohamed Ettaouil · 2017
In this paper, we propose a novel vision of image restoration problem. Based on the classical image restoration model, we introduce a modified optimization model with a new decision vector that indicates the pixels categories. Depending on the degraded image and its median corrected version, the technique allows via the optimization criterion to choose on which one of the both generates the best solution of the model, which in turn is solved by the genetic algorithm. Once noisy pixels are detected, a median based filter is performed only for these pixels. Experiments shows that the results are satisfactory in term of both visual quality and quantitative measurement.