comp.lang.idl-pvwave archive: archive » Re: KMean Clustering of RGB Images

Home » Public Forums » archive » Re: KMean Clustering of RGB Images

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: KMean Clustering of RGB Images [message #53902 is a reply to message #53901]

Wed, 09 May 2007 04:36

Mort Canty
Messages: 134
Registered: March 2003

Senior Member

helaha@gmx.net schrieb:

> Thank you Mort,
> concerning the "order of data points" it is necessary to think about
> the construction of the RGB Data space. The original RGB image is a 3
> dim. array e.g. [3,ImageSizeX, ImageSizeY]. The RGB Data space, which
> should be clustered concerning only the grey values and without
> consideration of the pixel locations, has two dimensions e.g. [3,
> NumberOfPixels ]. Therefore the individual image pixels are
> transformed in a sequence of RGB triple values ("data points"). The
> first triple values correspond to the pixel located at [3, 0, 0] e.g.
> the left bottom corner of the image. If the image is mirrored, rotated
> ore transposed before RGB space construction, then the first RGB
> triple value will be altered. Nevertheless all the individual image
> pixels were transformed into the RGB space, only the sequence is
> changed.
>
> Thanks,
> Helmut
>
>

Hi Helmut,

Now I see what you were getting at. But I think you're confusing things
a bit. K-means clustering takes place entirely in the feature space
(3-D RGB space in your case) and makes no reference to any ordering of
the pixels, spatial or otherwise. The Euclidean distances from each
pixel to each cluster mean are determined and the pixel is assigned to
the cluster for which the distance is smallest. Then the means are
recalculated and the procedure repeated until the means cease to change.
The order in which the distances are calculated in each iteration
clearly plays no role. The indeterminacy arises from the random
initialization of the cluster means.

Cheers

Mort

Report message to a moderator

[Message index]

		Re: KMean Clustering of RGB Images By: Mort Canty on Wed, 09 May 2007 07:43
		Re: KMean Clustering of RGB Images By: helaha on Wed, 09 May 2007 05:08
		Re: KMean Clustering of RGB Images By: Mort Canty on Wed, 09 May 2007 04:36
		Re: KMean Clustering of RGB Images By: helaha on Wed, 09 May 2007 03:00
		Re: KMean Clustering of RGB Images By: Mort Canty on Tue, 08 May 2007 06:13

Previous Topic:	Place a colorbar under each plot whose position is determined by !p.multi
Next Topic:	Re: Place a colorbar under each plot whose position is determined by !p.multi

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Fri Oct 10 20:01:52 PDT 2025

Total time taken to generate the page: 0.48083 seconds