Prev: shadow removal for traffic sequences
Next: Solve a LMI problem with some variables?(Please help)
From: Rui Xavier on 30 Jan 2010 02:23
hi!..im doing a project on ear recognition...iv got to find out the centroid of a given edge detected image...which i dont know the formulas or the steps to be carried out..m new to matlab and in urgent need of help....pls help
From: James Tursa on 30 Jan 2010 03:19
"Rui Xavier" <ruixavi(a)yahoo.com> wrote in message <hk0moo$pg6$1(a)fred.mathworks.com>...
> hi!..im doing a project on ear recognition...

.... Ah, that explains the yelling ...


































iv got to find out the centroid of a given edge detected image...which i dont know the formulas or the steps to be carried out..m new to matlab and in urgent need of help....pls help
From: ImageAnalyst on 30 Jan 2010 07:02
On Jan 30, 2:23 am, "Rui Xavier" <ruix...(a)yahoo.com> wrote:
> hi!..im doing a project on ear recognition...iv got to find out the centroid of a given edge detected image...which i dont know the formulas or the steps to be carried out..m new to matlab and in urgent need of help....pls help

---------------------------------------------------------------
Here, this will get you started:
http://www.mathworks.com/matlabcentral/fileexchange/25157

For your case you might have to try this:
do an edge detection
use imclose() to fill in gaps
use imfill to make the objects solid,
then call bwlabel (or bwconncomp)
then call regionprops.
From: Rui Xavier on 12 Feb 2010 13:54
hi!!..i appreciate u helping me in my search for the centroid of my ear image...however t code u have given calculates many centroids...which is not wat i am lookin for....i want jus the single centroid of the ear image...pls do help as quickly as u can because i have been stuck tryin to figure this out for a long time...tell me if there is the need to send u the ear images which i duly will ...waitin for replies..thanx
From: ImageAnalyst on 12 Feb 2010 14:23
"Rui Xavier"
The demo code finds the centroids of all objects that are in your
image. If you have a single object, you'll get a single centroid. My
guess is that you really haven't done a good job of finding your
single object - you don't really have a single object yet, do you? So
you need to do some more image processing to throw away all objects
you don't want (like the code shows) and keep just the one you do
want. It should be easy to adapt the m-code, assuming your "given
edge detected image" is a good one (I have my doubts about that
though). So maybe I can take a look at it this weekend if you'll
upload your photos to http://drop.io. Upload the original image, and
the edge detected version of it.
 |  Next  |  Last
Pages: 1 2 3 4
Prev: shadow removal for traffic sequences
Next: Solve a LMI problem with some variables?(Please help)