[Insight-users] Mask boundary to ordered set of points

Radhika Sivaramakrishna radhika.sivaramakrishna at synarc.com
Thu Nov 4 13:15:52 EST 2004


Hi Benjamin and Raghu,
Thanks for both your suggestions. These are very helpful.
I can get the seed point on the edge (in a similar way as Benjamin
suggested). And my region is very simple with no holes, and it is a single region. Given this starting point on the edge and assuming a simple single region, then is there already an ITK function available to get the ordered list of points. Raghu, is your tool implemented with ITK filters?
Radhika



-----Original Message-----
From: Raghu Venkatram [mailto:raghu.venkatram at gmail.com] 
Sent: Thursday, November 04, 2004 7:37 AM
To: king.benjamin at mh-hannover.de
Cc: insight-users at itk.org; Radhika Sivaramakrishna
Subject: Re: [Insight-users] Mask boundary to ordered set of points

Hi Radhika,

I did something similar couple of months back
http://ct.radiology.uiowa.edu/~jiangm/courses/dip/html/node126.html

the above link gives a very simnple boundary tracing algorithm, use
the 8 connected method.

I am assuming that you have already segmented and done some edge
detection. I used the above method to get teh boundary points, to do
this you need a seed point on your edge.

Thanks,
Raghu



On Thu, 4 Nov 2004 11:00:34 +0100, Benjamin King
<king.benjamin at mh-hannover.de> wrote:
> Hi Radhika,
> 
> I needed such a thing and made it myself als follows:
> 
> 1) Find bounding box of the object, cut with itk::RegionOfInterestImageFilter
> 2) Erode
> 3) Get border pixels by subtracting the erosion from the original
> 4) Iterate the result image and push all points in a std::vector
> 
> I iterated the whole region, but probably you can use a variant of
> itk::FloodFilledFunctionConditionalConstIterator as well.
> 
> > Or is there any function, where if you start from a boundary point, you
> > can move in a specified direction (clockwise or counterclockwise) and
> > get an ordered set of indices visited?
> This requires a border that is equivalent to a circle. Only two-dimensional
> objects with just one component, no holes and no thin (i.e. one pixel wide)
> areas meet this requirement. I don't think, there is a special class in ITK
> for those.
> 
> Hope it helps,
>   Benjamin
> 
> --
> Benjamin King
> Institut für Medizinische Informatik
> Medizinische Hochschule Hannover
> Tel.: +49  511  532-2663
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
>
 
--------------------------------------------------------

This email message is for the sole use of the intended recipient(s) and may contain confidential and privileged information.Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. If you are the intended recipient, please be advised that the content of this message is subject to access, review and disclosure by the sender's Email System Administrator.
--------------------------------------------------------


More information about the Insight-users mailing list