| KEYWORD |
Area Engineering
Compressed sensing
Reference persons ENRICO MAGLI
Research Groups Image Processing Lab (IPL)
Thesis type RESEARCH
Description Is is possible to sample signals below the Nyquist rate? The new field of "compressed sensing" shows that compressible signals (i.e., signals that are sparse in some domain, e.g. the wavelet transform) can be sampled well below the Nyquist rate. In other terms, the signals can be reconstructed from a very small number of incoherent measurements, obtained e.g. by taking the scalar product of the signal with some pseudorandom functions. This field is in its infancy, and the image/video applications require a lot of this research work. Thus, theses are available to apply compressive sampling to realistic signals
In particular, available theses will address one or more of the following research problems:
Improved signal models for compressed sensing of images and video sequences.
Improved image/video reconstruction from samples using image restoration techniques.
Distributed compressed sensing reconstruction algorithms
User localization based on compressed sensing
And many more topics that exploit compressed sensing to improve the performance of several signal processing tasks.
Note: *some* of the theses may have a significant mathematical content. Others are mor esignal processing or implementation oriented.
These theses are in the framework of project "CRISP - Towards Compressive Information Processing Systems" (www.crisp-erc.eu). This is an ERC projects, awarded to investigators of proven scientific excellence, to carry out frontier research on important topics.
Required skills Signal Processing skills, C/C++ programming.
Deadline 07/10/2017
PROPONI LA TUA CANDIDATURA