Image denoising based on a variable spatially exponent PDE (Q6144897)

This interesting paper studies the subject of multiplicative noise models which have many applications in medicine, radar, computer vision and many others. The environmental conditions of the corresponding image acquisition processes often make standard additive noise models difficult for the denoising of such images. For this reason, multiplicative noise models often give a better representation of consistent imaging patterns. The authors build a model for image denoising that has been contaminated by multiplicative noise, specially a Speckle one. They then build a class of partial differential equations whose nonlinear structure depends on a spatially tensor-depending quantity attached to the desired solution, which takes into account the gray-level information by introducing a gray-level indicator function in the diffusion coefficient. The authors then provide some discretization, theoretical results and a stability condition for the model built. Finally, the authors provide some numerical results to show the effectiveness of their model by comparing their results with some competitive models. The paper is well written with a good set of references.