Non-local Means Denoising using An Adaptive Kernel
Keywords:
Non-local means, Denoising, Shape adaptivepatches, Adaptive kernel.
Abstract:
Non-local means algorithm is one of the powerfulimage denoising methods. Maintaining noise near edges andtextural parts of a noisy image, is one of the main drawbacks ofNLM. In this paper we introduce an adaptive kernel derivedfrom image structure to remove maintained noise. Experimentalresults show superiority of our algorithm in comparison withoriginal NLM as well as a method based on shape adaptivepatches.