Rank: 7983 based on 7 downloads (last 30 days) and 1 file submitted
photo

David Provencher

E-mail
Company/University
Université de Sherbrooke

Personal Profile:
Professional Interests:
signal processing, MRI, fMRI, optical imaging, electroencephalography

 

Watch this Author's files

 

Files Posted by David
Updated   File Tags Downloads
(last 30 days)
Comments Rating
07 Aug 2014 lsqnonnegvect.m vectorized lsqnonneg Author: David Provencher lqnnonneg, nonnegative leastsqua... 7 0
Comments and Ratings by David View all
Updated File Comments Rating
08 Aug 2014 EM algorithm for Gaussian mixture model with background noise Standard EM algorithm to fit a GMM with the (optional) consideration of background noise. Author: Andrew

I'm trying to run the code, but I keep getting this warning :

'Warning: chol failed, algorithm abandoned';

because the cholcov(Sigma(:,:,j),0); line always fails at the 2nd iteration (bn_noise='T') or 3rd iteration (bn_noise='F').

FYI, I have no NaN values in my data, and I get coherent results with kmeans() and emgm() [the submission that inspired this one]. Actually, no matter what data I feed into the function (e.g. squre matrix, rand(m,n), ...) this step always fails.

Any insight on this?
Thanks,
David

22 Mar 2013 Compare Structures Compares two structured variables recursively and notes where the two structures are different. Author: michael arant

Very useful. Would have liked a 'silent' option to disable output to command window and use only the output variables for large structures.

05 Dec 2011 inpaint_nans Interpolates (& extrapolates) NaN elements in a 2d array. Author: John D'Errico

Ok, I suspected this might be the case. I was already doing what you suggest, but I figured I might ask just in case.

Thanks for your answer!

01 Dec 2011 inpaint_nans Interpolates (& extrapolates) NaN elements in a 2d array. Author: John D'Errico

Works great!

I'm wondering if there would be a way to exclude some NaNs from the interpolation (to keep them as NaNs). At the moment, I'm masking the results away, but I'd rather exclude them altogether. I thought about modifying nan_list and known_list, but it seems to me I'd run onto problems if the input matrix wasn't "full". Any help on this?
Thanks

(I'm using inpaint-nans in a stereo vision application to interpolate holes in the disparity maps. However, the images from which I compute disparity maps may have very large regions of invalid pixels which I don't need/want to interpolate).

19 Aug 2011 Stereo Matching A GUI demonstrates an efficient stereo matching algorithm. Author: Wim Abbeloos

There seems to be an error in the integral image technique. The filtering window should be

h = zeros(WS+2,WS+2,'double');
h(1,1) = 1; h(1,WS-1) = -1;
h(WS-1,1) = -1; h(WS-1,WS-1) = 1;

that is, for a 3x3 summing window, the filtering window must be 5x5 to implement the algorithm correclty (see http://en.wikipedia.org/wiki/Summed_area_table#The_algorithm and note that points A, B and C are outside the summing window)

I tested this with a simple matrix for validation. Otherwise, thanks for the useful code.

Contact us