[ITK-users] [ITK] help- find position of seed-levelset segmentation/connected treshold
Matt McCormick
matt.mccormick at kitware.com
Sat Jun 7 15:39:48 EDT 2014
Hi Sindoor,
itk::Image contains methods like TransformPhysicalPointToIndex [1]
that can be helpful.
Matt
[1] http://www.itk.org/Doxygen/html/classitk_1_1ImageBase.html#af4a7c9c3787e9fdafbaaade2e02efa25
On Sat, Jun 7, 2014 at 7:09 AM, Sindhoor Bhat <bhatsindhoor at ymail.com> wrote:
> Dear Matt/itk users,
> Now I'm able to pick world coordinates using vtk point picker widget.
> But for seed input to itk, we need index. I'm really struck. can you pls
> help me how to move further? using itk-vtk interaction for seed selection?
>
> Also, to call vtk window for point selection which command should i use? (is
> it call-back function in vtk? But from itk how to call ?)
>
> thank you
>
>
> On Friday, 6 June 2014 6:04 PM, Matt McCormick <matt.mccormick at kitware.com>
> wrote:
>
>
> Hi Sindhoor,
>
> In your previous mail, you described taking sub-regions, which changes
> index positions. To sample a value from a physical position, use an
> itk::InterpolateImageFunction.
>
> The apparent flip may be due to a difference in the direction and
> location of the Y-axis in your visualization programs.
>
> Hope this helps,
> Matt
>
> On Fri, Jun 6, 2014 at 1:03 AM, Sindhoor Bhat <bhatsindhoor at ymail.com>
> wrote:
>> Thanks Matt for your suggestion.
>> *But I've tried with that too. It is not considering the index/seed
>> position
>> that I write in program.
>> When I added below code to get selected seed position, it shows
>> completely different point than what I had written(using image J --->
>> physical coordinates )
>> std::cout << filter->GetOutput()->GetPixel( Position1 ) <<
>> std::endl;
>> *seed point doesn't appear between selected thresholds ( say: (85,300) &
>> seep position will be like 45 etc)
>>
>> *Also, I should mention that my dicom images appear in reverse order(
>> flipped). Is this affecting my seed selection?
>> (I tried subtracting Y coordinate from maxY etc.)
>>
>> Can you pls tell what is wrong? thank you.
>>
>
>
More information about the Insight-users
mailing list