No subject
Sun Apr 3 09:54:08 EDT 2011
segmentation filter (confidence connected to be specific) will construct a
binary image.
After I do that kind of segmentation, I view the binary image in vtk and
then I try to carry out the same segmentation filter on that binary image so
that I can fill up any black holes
lingering around in the image. I want to do this because of the easiness of
adding seeds with confidence connected filter.
However, I dont exactly get the result I want, what appears is just a white
window, all white.
Is it not possible to segment binary images, are there alternatives to get
rid of black holes in segmentation result where I can use my adding seed
interface quite easily. I'm using vtk to get
the user to click on the image to add their seeds and I was hoping that I
could do the same for those black holes.
Thanks,
John
--0015174a064088f29f04adc62255
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Hi ITK users,<div><br></div><div>From what I can understand, segment dicom =
images using a region growing segmentation filter (confidence connected to =
be specific) will construct a binary image.</div><div><br></div><div>After =
I do that kind of segmentation, I view the binary image in vtk and then I t=
ry to carry out the same segmentation filter on that binary image so that I=
can fill up any black holes</div>
<div>lingering around in the image. I want to do this because of the easine=
ss of adding seeds with confidence connected filter.</div><div><br></div><d=
iv>However, I dont exactly get the result I want, what appears is just a wh=
ite window, all white.</div>
<div><br></div><div>Is it not possible to segment binary images, are there =
alternatives to get rid of black holes in segmentation result where I can u=
se my adding seed interface quite easily. I'm =A0using vtk to get</div>
<div>the user to click on the image to add their seeds and I was hoping tha=
t I could do the same for those black holes.</div><div><br></div><div>Thank=
s,</div><div>John</div>
--0015174a064088f29f04adc62255--
More information about the Insight-users
mailing list