[Insight-users] Simple registration question (ImageRegistartion1.cxx )

[Luis Ibanez]

Hi Marco,

The resample filter requires you to provide
this default pixel value.

You could compute "similar" pixels values
to what your moving image have in the border,
but,.. you will have to do that computation
on your own (for example, by building a mask
and using the Statistics Label Map classes).

Another option is to let your subsequent
analysis know that the "default" pixel
values are not to be considered.

and

Another option is to "fill up" the empty
spaces of the moving image with values
taken from the corresponding pixels
of the fixed image (this is just for
cosmetic purposes, and it is technically
an incorrect way of completing the image).


    Regards,


          Luis

------------------------------------------------------
On Wed, Sep 29, 2010 at 3:56 AM, marco giordano <marco.giord at gmail.com>wrote:

>
> Thank you. I see also what is the cause...
>
> Is there a way to avoid setting the defaultFixelValue to 100 and set it in
> a way to complete the moving image (e.g. complete the image using the old
> values of the moving image or some interpolated version of that). I really
> do not have clear if this would be appropriate but I would like not to have
> empty/flat spaces in my registered image because they make my slices
> unusable for analysis.
>
> I hope I explained me clear.
>
> Thx
>
> 2010/9/28 Navneeth Subramanian <navneeth.s at gmail.com>
>
> The cause is seen in line 600:
>> resampler->SetDefaultPixelValue( 100 );
>>
>> This is the value being applied during the interpolation of the moving
>> image based on the computed transform.
>> Change it & you're all set.
>>
>> best,
>> navneeth
>>
>>
>> On Sat, Sep 25, 2010 at 1:13 AM, marco giordano <marco.giord at gmail.com>wrote:
>>
>>> Yes, exactly the last slice of the registered image.
>>>
>>> Is there anyway to limit this effect?
>>>
>>> What if one adds empty slices on the lowest (highest) border? does that
>>> affect the registration?
>>>
>>> Thx
>>>
>>>  2010/9/24 Kishore Mosaliganti <kishoreraom at gmail.com>
>>>
>>> By last image, did you mean the last slice of the registered (moving)
>>>> image?
>>>>
>>>> If so, I think this is because of the image transform applied that
>>>> moves the image totally out of that image plane. It will happen to the
>>>> boundary slices usually.
>>>>
>>>> Kishore
>>>>
>>>> On Fri, Sep 24, 2010 at 10:56 AM, marco giordano <marco.giord at gmail.com>
>>>> wrote:
>>>> > Hi,
>>>> >
>>>> > I have a simple issue that I can not solve neither understand the
>>>> cause of
>>>> > it.
>>>> >
>>>> > I run the Example ImageRegistartion1.cxx on 2 3D CT images
>>>> (256x256x11)
>>>> > modyfing the code to take into account the third dimension and float
>>>> number
>>>> > for representation.
>>>> >
>>>> > Output looks like this:
>>>> >
>>>> > ....
>>>> > 25 = 104.112 : [0.097946, -0.00159973, 0.00035025]
>>>> > 26 = 104.082 : [0.0979543, -0.00155174, 0.000494857]
>>>> > Result =
>>>> >  Translation X = 0.0979543
>>>> >  Translation Y = -0.00155174
>>>> >  Translation Z = 0.000494857
>>>> >  Iterations    = 28
>>>> >  Metric value  = 104.135
>>>> >
>>>> > Registration runs succeffully but I notice in my registrated image
>>>> that the
>>>> > last slice (slice Number 11) is all set to the Deafult Value 100.
>>>> >
>>>> > Is that due to the interpolation or something I did not specify in my
>>>> code?
>>>> >
>>>> > Is it possible to avoid this effect?
>>>> >
>>>> > Thank you very much.
>>>> >
>>>> >
>>>> > --
>>>> > M. Giordano
>>>> >
>>>> >
>>>> > _____________________________________
>>>> > Powered by www.kitware.com
>>>> >
>>>> > Visit other Kitware open-source projects at
>>>> > http://www.kitware.com/opensource/opensource.html
>>>> >
>>>> > Kitware offers ITK Training Courses, for more information visit:
>>>> > http://www.kitware.com/products/protraining.html
>>>> >
>>>> > Please keep messages on-topic and check the ITK FAQ at:
>>>> > http://www.itk.org/Wiki/ITK_FAQ
>>>> >
>>>> > Follow this link to subscribe/unsubscribe:
>>>> > http://www.itk.org/mailman/listinfo/insight-users
>>>> >
>>>> >
>>>>
>>>
>>>
>>>
>>> --
>>> Marco Giordano
>>>
>>>
>>> _____________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Kitware offers ITK Training Courses, for more information visit:
>>> http://www.kitware.com/products/protraining.html
>>>
>>> Please keep messages on-topic and check the ITK FAQ at:
>>> http://www.itk.org/Wiki/ITK_FAQ
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://www.itk.org/mailman/listinfo/insight-users
>>>
>>>
>>
>
>
> --
> Marco Giordano
> MSN:gilmour812002 at yahoo.it <MSN%3Agilmour812002 at yahoo.it>
> ICQ :285-118-610
> SKYPE:marcogiord81
>
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Kitware offers ITK Training Courses, for more information visit:
> http://www.kitware.com/products/protraining.html
>
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20101017/2ba1f293/attachment.htm>