A Wavelet-TV approach to super resolution of MRI data.


Supervisor
Matteo Frigo - EPFL & Master degree in Mathematics Verona

Date and time
Friday, February 17, 2017 at 10:30 AM - laboratorio ALFA

Contact person

Publication date
February 2, 2017

Department
Computer Science  

Summary

Super resolution (SR) is the science of enhancing the 
resolution of an image in a coherent way. Medical imaging (Radiography, 
MRI, Ultrasound, Tomography ... ) is maybe the best customer of this 
field of research since a higher spatial resolution makes the diagnosis 
easier and more accurate. SR has been faced from the optical/acquisition 
side and from the geometrical/post-processing side and in this talk we 
are going to present a technique coming from the latter category. We 
will exploit the multi-resolution features of the Wavelet transform to 
adapt to the local characteristics of the image, then we will see how 
Total Variation provides important local informations which will be used 
to reconstruct the subpixel information that we were seeking.

Keywords: Super Resolution, Wavelet Transform, Total Variation, Medical 
Imaging

© 2002 - 2022  Verona University
Via dell'Artigliere 8, 37129 Verona  |  P. I.V.A. 01541040232  |  C. FISCALE 93009870234