From: subrajeet Mohapatra on
Hi,

I have a blood microscope image from which i got the nucleus separated using K means but now i want the black background to be removed and it shd be white so that i can use sub imaging for feature extraction.
So how it can be done.

image link is
http://drop.io/957b1yy

Thanks in advance
From: ImageAnalyst on
imageArray(imageArray == 0) = 255;
From: subrajeet Mohapatra on
ImageAnalyst <imageanalyst(a)mailinator.com> wrote in message <a3cbab1d-e44d-4e9a-a6d9-8af4b8c37ef7(a)j27g2000vbp.googlegroups.com>...
> imageArray(imageArray == 0) = 255;

One more doubt
In this image can I remove the background I mean I need the whole violet part not its neighboring light yellowish part.

Images are
http://drop.io/np0ihon2011


Actually I applied the kmeans to segment but loosing some informations
From: ImageAnalyst on
Yes, you'll have to detect a color (yellow or purple).
See my color detection example:
http://www.mathworks.com/matlabcentral/fileexchange/authors/31862
Then you may want/need to do some cleanup such as filling holes or
throwing away small noise blobs.

kmeans should be able to do it also like in this demo:
http://www.mathworks.com/products/demos/image/color_seg_k/ipexhistology.html
but I don't have the Statistics Toolbox so I can't try it or help you
with it.