[vtkusers] vtkImageErode3D

elena bresciani elena.bresciani87 at gmail.com
Thu Nov 28 03:34:38 EST 2013


I tried to set different values of kernel size and erode/dilate values
without seeing any difference in the output; also I added the Update() call
and tried using he two filters separately, still no changes.

here's my code (I'm using python):

# imageReader is a vtkBMPReader

# apply the dilate/erode filter to the image

erode = vtk.vtkImageDilateErode3D()
erode.SetInputConnection(imageReader.GetOutputPort())
erode.SetErodeValue(3)
erode.SetKernelSize(5, 5, 1)
erode.ReleaseDataFlagOff()
erode.Update()

dilate = vtk.vtkImageDilateErode3D()
dilate.SetInputConnection(erode.GetOutputPort())
dilate.SetDilateValue(3)
dilate.SetKernelSize(5, 5, 1)
dilate.ReleaseDataFlagOff()
dilate.Update()

# create the actors one for the original image ad one for the
# processed one
originalActor = vtk.vtkImageActor()
originalActor.GetMapper().SetInputConnection(imageReader.GetOutputPort())

processedActor = vtk.vtkImageActor()
processedActor.GetMapper().SetInputConnection(dilate.GetOutputPort())

# rendering stuff...


2013/11/27 Jon Haitz Legarreta <jhlegarreta at vicomtech.org>

> Dear Elena,
> another alternative would be to vary the kernel size to some extreme
> values so that you can really be aware of any changes.
> Another possibility would be to use the erode/dilate filters separately.
> And yes, you are right, if you erode an object to the ground it cannot be
> dilated anymore however hard you try.
>
> HTH,
> JON HAITZ
>
>
>
> On 27 November 2013 13:29, Marco Nawijn <nawijn at gmail.com> wrote:
>
>> Hi Elena,
>>
>> Although I probably cannot help you directly, I would advise you
>> isolate your problem to the smallest compilable piece of code
>> and post this so experts can take a look at it. One thing that
>> bit me in the past is that ones you apply a filter, you have to
>> a.) call Update() on it and b.) call GetOutput() to get the result
>> you are looking for. Without seeing your code however it is
>> difficult to judge.
>>
>> Kind regards,
>>
>> Marco
>>
>> P.S. There is nothing wrong with your english!
>>
>>
>>
>>  On Wed, Nov 27, 2013 at 12:24 PM, elena bresciani <
>> elena.bresciani87 at gmail.com> wrote:
>>
>>>  I'm getting started with vtk and I'm still unable to solve a lot of
>>> problems.
>>>
>>> I'm trying to apply the vtk filter vtkImageDilateErode3D to a CT image
>>> but I
>>> can't obtain the result I'm looking for.
>>> In fact I expect the result to be a at least slightly different image but
>>> the output is instead an image identical to the one given as input (I
>>> even
>>> checked subtracting the two images with vtkImageMathematics and after the
>>> operation there's really nothing left!).
>>> I don't understand if the problem is in the setting of the Erode and
>>> Dilate
>>> values and in the kernel size or maybe in the function of the filter
>>> itself,
>>> i.e. I expect that the output image has something less compared to the
>>> original one after erosion and dilatation (what has been eroded "to the
>>> ground" canot be dilated back - is this correct?)
>>>
>>> Thank you in advance
>>> Sorry for the bad english
>>>
>>> Elena
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Please keep messages on-topic and check the VTK FAQ at:
>>> http://www.vtk.org/Wiki/VTK_FAQ
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://www.vtk.org/mailman/listinfo/vtkusers
>>>
>>>
>>
>> _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the VTK FAQ at:
>> http://www.vtk.org/Wiki/VTK_FAQ
>>
>> Follow this link to subscribe/unsubscribe:
>> http://www.vtk.org/mailman/listinfo/vtkusers
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.vtk.org/pipermail/vtkusers/attachments/20131128/fb2a1289/attachment.htm>


More information about the vtkusers mailing list