Dictionary Learning for Real Neuro Data of the Visual Cortex

Tutor: Tobias Schnier
Type of Thesis: Project (MSc)
date of end: 04/2016
Student: Maik Röper
Status: finished

In this project possible dictionaries for real neuro data are evaluated and new dicitionaries proposed.

Main questions:

Do standard bases like Fourier or Wavelet describe Neuro data sufficiently?

Will using Dictionary Learning algorithms lead to better results?


Through wireless implants neuro data is transmitted out of the brain via Compressed Sensing (CS) algorithms. To reconstruct them properly a well designed dictionary is needed that can represent the real neuro data in a practical way.

Work to do:

Literature search of common dictionaries

Test found dictionaries on real neuro data

Literature search of dictionary learning algorithms

MATLAB-programming of found dictionary learning algorithms and testing them on real neuro data

Expertise needed

CS knowledge is helpful but not a requirement

Basic mathematical background is needed

Basis MATLAB knowledge is needed

Last change on 31.05.2016 by
AIT ieee GOC tzi ith Fachbereich 1
© Department of Communications Engineering - University of BremenImprint / Contact