Re: ImageJ PCA

Posted by Gabriel Landini on Apr 13, 2007; 3:46pm
URL: http://imagej.273.s1.nabble.com/ImageJ-PCA-tp3699745p3699752.html

On Friday 13 April 2007 14:38:40 aszia787 wrote:
> My question is, is the first eigenimage that we
> get, is that the eigneimage of image A or image B? i guess it doesn't tell
> me that the first image returned is the eigen image of which original
> feature/image as i have to add the original features with the maximum
> eigenvalue.

I am no expert in PCA , but as I understand it, the plugin calculates new
combinations of the original images that explain the variance of the data and
puts them in decreasing order of "variance explanation". That is returned, I
believe, in the eigenvalues spectrum graph (but please correct me if I am in
the wrong track).

I do not think that the plugin gives you information about "what component
corresponds to mixtures of which images" like is reported in some stats
programs like SPSS or Minitab.

There is some information with references here (but perhaps you already know
this):

http://bij.isi.uu.nl/pca.htm

Have you consulted Michael Abràmoff (the author of the plugin)?

Cheers,

G.