matlab: optimizing code for computing statistics in multi-scale circular neighborhoods

Question

I\'ve a stack of images (imgstack) over which I would like to compute some statistics (e.g. mean, std, median) in multi-scale circular neighborhood

Answer 1

You can replace all operations using filters, which should be significantly faster.
For each imagestack_radii, first create a circular mask:

n = getnhood( strel('disk', imagestack_radii(s), 0 ) );

• mean: use imfilter with double(n)/sum(n(:)) as filter

  imgstack_feats{ss}(:,:,1) = imfilter( imgstack(:,:,ss), double(n)/sum(n(:)), 'symmetric' );
  

• std: once you have the mean, you can compute the second moment using

  tmp = imfilter( imgstack(:,:,ss).^2, double(n)/sum(n(:)), 'symmetric' );
  imgstack_feats{ss}(:,:,2) = sqrt( tmp - imgstack_feats{ss}(:,:,1) );
  

• median: use ordfilt2

  imgstack_feats{ss}(:,:,3) = ordfilt2( imgstack(:,:,ss), ceil( sum(n(:))/2 ), n );