Abstract : This work proposes two methods for MRI denoising that exploit both the sparseness and self-similarity properties of the image. The proposed methods are based on 1) an overcomplete non-local patch grouping PCA hard thresholding and 2) a rotationally invariant nonlocal means filter that uses the output of first filter as a guide image to filter the image.
https://hal.archives-ouvertes.fr/hal-00961154 Contributor : Pierrick CoupéConnect in order to contact the contributor Submitted on : Wednesday, March 19, 2014 - 3:18:56 PM Last modification on : Monday, December 20, 2021 - 4:50:11 PM
Jose Vicente Manjon, Pierrick Coupé, Antoni Buades. Non-local PCA-based MRI denoising. Organization for Human Brain Mapping 2014 Annual Meeting, Jun 2014, Hamburg, Germany. ⟨hal-00961154⟩