VisAGeS Online Applications Page

Data process configuration

Process selection

The method proposed in this paper is based on an optimized version of the Non Local (NL) Means algorithm.
This approach uses the natural redundancy of information in image to remove the noise.

See "P. Coupe, P. Yger, S. Prima, P. Hellier, C. Kervrann and C. Barillot. An Optimized Blockwise Non Local Means Denoising Filter for 3D Magnetic Resonance Images. Accepted in Transactions on Medical Imaging".(pdf/html)

Contact Pierrick COUPE for more details on this denoising method.

Please contact Alexandre ABADIE if you have problems when using the application.

This filter is an improvement of the Non Local (NL) Means filter for 2D image denoising.

See Ch. Kervrann, J. Boulanger, P. Coupe. Bayesian non-local means filter, image redundancy and adaptive dictionaries for noise removal.
In Proc. Conf. Scale-Space and Variational Meth. (SSVM' 07), pages 520-532, Ischia, Italy, June 2007.
See also the demo web page.

Contact Charles Kervrann for more details.

Please contact Alexandre ABADIE if you have problems when using the application.

This the registration process : you have to upload 2 volumes, one as the reference and the other as the moving.
Ask Nicolas Wiest-Daessle for more details on this registration method.

Please contact Alexandre ABADIE if you have problems when using the application.

This process takes a 3D grey value image as an input, computes the mid-sagittal plane (MSP) of the bilateral object in the image (typically a human brain), and outputs an image in which the computed MSP is aligned with the plane in the centre of the image grid.

Ref: S. Prima, N. Wiest-Daessle. Computation of the mid-sagittal plane in diffusion tensor MR brain images
In Proceedings of SPIE Medical Imaging 2007: Image Processing, I. Taylor (ed.), San Diego, USA, February 2007.

Contact and for more details on this process.

Please contact Alexandre ABADIE if you have problems when using the application.

Fill in your rician denoising parameters Fully automated mode Expert mode

PARAMETERS :

Smoothing parameter h =

Neighboring size is (2d+1)^3, d =

Search volume size is (2M+1)^3, M =

VOXEL SELECTION :

Test on mean : No Yes Threshold selection mu [0 - 1] :

Test on variance : No Yes Threshold selection sigma [0 - 1] :

IMPLEMENTATION :

Voxelwise : Blockwise :

Fill in your denoising parameters Fully automated mode Expert mode

PARAMETERS :

Noise standard deviation =

Patch size =

Search window size =

Iteration number

Spacial sub-sampling

Fill in your registration parameters Auto Expert

Registration type :

Algorithm :

Cost function :

Joint histogramm interpolation :

Final interpolation method :

Fill in your symmetry plane parameters Fully automated mode Expert mode

PARAMETERS : None given for moment => execution in fully automated mode

Select data to process

Data format

Fill in your data format parameters Lines : Columns : Slices : Timeslices :

dx : dy : dz : dt :

Data type :

Orientation :

Select original data

Header :

Data :

Be careful you can only load data up to 50 Mo

Select floating data to register

Header :

Data :

Be careful you can only load data up to 50 Mo

Select data to process

Data format

Image :

Email for receiving results : Email :

Welcome to the VisAGeS online applications page