GPGPU based implementation of 3D Non local means Filter

he Non-local means denoising filter has been established as gold standard for image denoising problem in general and particularly
in medical imaging due to its efficiency. However, its computation time limited its applications in real world application. In this paper,
a distributed version on parallel hybrid architecture is proposed to solve the computation time problem, a new manner to computer
the filters’ coefficients is also proposed. The obtained results demonstrate the efficiency of the proposed method. Moreover, the
implementation is compared to that of other techniques, recently published in the literature, on standard benchmarks
he Non-local means denoising filter has been established as gold standard for image denoising problem in general and particularly
in medical imaging due to its efficiency. However, its computation time limited its applications in real world application. In this paper,
a distributed version on parallel hybrid architecture is proposed to solve the computation time problem, a new manner to computer
the filters’ coefficients is also proposed. The obtained results demonstrate the efficiency of the proposed method. Moreover, the

 

The Non-local means denoising filter has been established as gold standard for image denoising problem in general and particularly in medical imaging due to its efficiency. However, its computation time limited its applications in real world application. In this work, a distributed version on parallel hybrid architecture is proposed to solve the computation time problem, a new manner to computer the filters' coefficients is also added.

The obtained results demonstrate the efficiency of the proposed method where we achieved a speed up > 500.

This software was done in collaboration with the company BULL.

 

Source code [link]

CerebMov

This soft results from a collaboration between the "Antenne Analyse et Restauration du Mouvement (Paristech-ENSAM CNRS 8005) of Centre Hospitalier Universitaire Henri Mondor de Créteil" and the academic research laboratory LiSSi (Laboratoire Images, Signaux et Systèmes Intelligents) of Université Paris Est Creteil.

 

Download the version 1.2: 

It runs on windows xp or vista 32 

Matlab MCR->here

Dowload the exe -> here

 

For more details about the project click on this (Link).