Prev: www.jerseysuppliers.com Cheap NFL Jerseys Cheap NHL Jerseys Cheap MLB Jerseys
Next: debugging c++ dll using libmx.dll
From: Vassilis on 9 May 2010 03:34
Hi all,
I wonder about the usefulness of the "skeleton" and "the thinnest version" of an object in a binary image.
Suppose we have binary handwritten letters and we want to split them at the junction points.
I think that the following should work well
Thin_A=bwmorph(A,'thin',Inf);
Cross1_A=imfilter(double(Thin_A),ones(3))>3;
Cross_A=Cross1_A & Thin_A;

But i have to be sure that the result of thinning "preferes" 8-connectivity
(as this example)
i.e. for this subimage
0 1 0 1 1
0 1 1 0 1
0 1 0 1 1
the result is
0 1 0 1 0
0 0 1 0 1
0 1 0 1 0

Is the 'skel' useful for this problem? I think, it isn't.
Actuallly, does anybody know an application for which the 'skel' option works better than 'thin'?
And another question, wher can i find the algorithm used for skeletons?
 | 
Pages: 1
Prev: www.jerseysuppliers.com Cheap NFL Jerseys Cheap NHL Jerseys Cheap MLB Jerseys
Next: debugging c++ dll using libmx.dll