Hi

I suggest you that use from these items at this address,those were very useful for me.

www.coursera.org

and this course

''Image and video processing  From Mars to Hollywood''

at ''Week 8 Sparse modeling and compressed sensing''

good luck

Hi Rohit,

I wrote tutorial of OMP for myself when I learned compressive sensing.

You can have a look at my site : kore76.wordpress.com  under TUTORIAL Section.

After understanding OMP, I think it is easy to implement it in MATLAB. Let me know if you have question.

Good luck!

Hi Mr. Rohit,

I will attach the code below.

Don't hesitate to contact me if you have some comments,

%%  A is the N x D predefined dictionary. 

%%  w is the sparse vector

w = zeros(D,1);

NoOfIterations = 5;

for ii = 1:NoOfIterations

   r = y - A*w;

   g  = A'*r;

   temp = find(abs(g) == max(abs(g)));

   test = find(temp(1) == index);

   if(isempty(test) == 1)

       index = [index, temp(1)];

   end

   w(index,1) = inv(A(:,index)'*A(:,index))*A(:,index)'*y;

   r = y - A*w;

   s(:,ii) = w;

end

Hi,

Please find the MATLAB code attached.

BR,

Ahmad.

Dear Mr. Ahmed Bazzi

thanks fro your code, but i have question about No of Iterations, how can i determined this number according to the input values

Dear Mr. Ahmed Bazzi,

thank you

You may like to look at my tutorials for pursuit algorithms here https://sparse-plex.readthedocs.io/en/latest/book/pursuit/omp/algorithm.html .

I build the discussion step by step from basic ideas for pursuit algorithms, moving on to matching pursuit and then delving into OMP.

There is a hands-on section on OMP which guides through the execution of OMP on a simple problem step by step https://sparse-plex.readthedocs.io/en/latest/book/pursuit/omp/algorithm.html#hands-on-with-orthogonal-matching-pursuit.

I also provide an extremely fast implementation of OMP which is discussed here https://sparse-plex.readthedocs.io/en/latest/book/pursuit/omp/fast_omp.html

Ahmad Bazzi kindly tell me about A,x and q in this code. what these variables hold?