[Insight-users] problem with Chan and Vese method of segmentation

Juan Cardelino juan.cardelino at gmail.com
Fri Jan 28 15:00:59 EST 2011 

Hello Kishore,
             Thanks for your answer. I think I have now a clear
understanding of the situation. I certainly want to contribute to ITK,
but I need to get a deeper understanding of the level set framework
before contributing with code. I will do this while trying to
re-implement my code using ITK. In the meantime, I could review
documentation and provide examples (the ones I've made for myself) in
the places they are needed.

Another question about the Chan-Vese approach: I understand that it is
possible to run it on vector-valued images, but is it straightforward
to convert the scalar examples? Do you have any examples for that?


Thanks for your time.
Best,
       Juan

On Thu, Jan 27, 2011 at 6:47 PM, Kishore Mosaliganti
<kishoreraom at gmail.com> wrote:
> Hi Juan,
>
> Thank you for the email.
>
> 1. This is a process issue with the Insight Journal. Hopefully, it
> will improve through the ITKv4 efforts (via the GIT repository).
>
> 2. One of the problems with the PDF update is that the person who
> submitted the article is only allowed to modify it. Outside users
> cannot add or modify the content. So, this process needs to be changed
> as well. This article can be copied to a wiki page that will allow
> users to update it.
>
> 3. We are now refactoring the level-set code in ITK for the v4
> release. You will see many changes soon. We will also improve
> documentation and provide many examples for users. If you would like
> to contribute in the code development or on the documentation aspect,
> that will be great.
>
> 4. The multiphase examples were placed in InsightoClopedia wiki page.
>
> HTH,
> Kishore
>
> On Thu, Jan 27, 2011 at 12:38 PM, Juan Cardelino
> <juan.cardelino at gmail.com> wrote:
>> Hey Kishore,
>>
>> Thank you very much for your answer,
>>
>> I've learned everything you tell me in this mail, the day after my
>> quests,  after long hours of digging around the  ITK source code. I
>> was able to run the examples successfully, but the Insight Journal
>> submission really misdirected me, and I didn't find that mistake until
>> I found a previous post from you on the list.
>>
>> That makes me wonder two things:
>> 1) why is that code still kept in the IJ site? It is not being
>> compiled against new versions of ITK so now it is kinda broken, and
>> somehow shadows the classes on the review code.
>> 2) Would it make sense to update the PDF on IJ to reflect this? I'm
>> not only asking, I could volunteer to do it. I mean, the names of the
>> examples and the parameter list doesn't match the ones on the IJ
>> submission, and that's an additional rock on the path of people trying
>> to learn about this filters.
>>
>> I think many good features of ITK (those not on the Software Guide)
>> suffer from visibility problems and lack of a clear starting point
>> from a newcomer. Which in turn reflects in many repetitive questions
>> being posted in the list. That triggers the 3rd question to me:
>>
>> 3) Is it possible, and of interest of the community, to add this kind
>> of documentation (like the PDF in the IJ) to the Software Guide?
>>
>> I had this same problem when trying to learn about the Geodesic Active
>> Contours with Shape Guidance, there were many obscure parts out there
>> that I had to fill in, and I would like to contribute.
>>
>>
>> Back to the Chan-Vese code, I have a few questions left:
>> 4) is there any examples of the multiphase version of the code? It is
>> not on the review examples folder for sure.
>>
>> I'm very interested on this because In my group we have our own level
>> set toolbox and I intend to port the interesting things to ITK. i.e.
>> one region based algorithm that I've made in my master's thesis, which
>> will add some features to the current chan-vese framework.
>>
>> Anyway, thanks for the answer and for your great contribution to our community.
>> Best regards,
>>                  Juan
>>
>> On Wed, Jan 19, 2011 at 4:26 PM, Kishore Mosaliganti
>> <kishoreraom at gmail.com> wrote:
>>> Hi all,
>>>
>>> There is *NO* need for borrowing the Insight Journal classes. Please
>>> make sure that you compile ITK with the Review option turned on.
>>>
>>> The level set filters were renamed and moved to the ITK/Code/Review folders.
>>>
>>> Two examples of using the dense and sparse field filters are located
>>> at ITK/Examples/Review/itkScalarSinglePhaseDense2DTest.cxx and
>>> ITK/Examples/Review/itkScalarSinglePhaseSparse2DTest.cxx.
>>>
>>> There are also numerous examples of its usage at
>>> ITK/Testing/Code/Review/. You can grep the relevant files in that
>>> folder.
>>>
>>> I am able to compile and use these filters with ITK4.0. If you face
>>> any further difficulties, please send me the error message.
>>>
>>> Kishore
>>>
>>> On Tue, Jan 18, 2011 at 6:57 AM, Juan Cardelino
>>> <juan.cardelino at gmail.com> wrote:
>>>> Hello Julien,
>>>> Where you been able to do solve this? I've been struggling with the
>>>> same issue without luck. I'm using ITK 3.19, and I get the following
>>>> results:
>>>> * If I remove everything but the main, the includes are not found.
>>>> * If I add them, it seems that they conflict somehow with the
>>>> installed ITK classes.
>>>> If someone is interested, I could upload the error I got.
>>>>
>>>> For the original authors,
>>>> I gave it a try with ITK 4.0 from GIT but there were some classes that
>>>> have been refactored (itkEuclideanDistance).  So I fell back to 3.19.
>>>> Are you planning to update this? or have you already done it?
>>>> In addition, would it be possible to enumerate explicitly which
>>>> classes are required and which not? It is really hard to tell at the
>>>> begining.
>>>>
>>>> Thanks in advance and great work.
>>>> Best regards,
>>>>                   Juan
>>>>
>>>>
>>>> On Wed, Jun 23, 2010 at 4:26 AM, julien_s <surchenjohn at gmail.com> wrote:
>>>>>
>>>>> I did a folder just with my .cxx et a CMakeList.txt but I have errors because
>>>>> my file .cxx include ".h" which were present in the Insight Journal Folders
>>>>> but not in the Review folder.
>>>>>
>>>>> For example : "itkSparseMultiphaseLevelSetImageFilter.h"
>>>>>
>>>>>
>>>>> (Maybe I also make some errros in the next CMakelist:
>>>>>
>>>>> # This is the root ITK CMakeLists file.
>>>>> CMAKE_MINIMUM_REQUIRED(VERSION 2.4)
>>>>> IF(COMMAND CMAKE_POLICY)
>>>>>  CMAKE_POLICY(SET CMP0003 NEW)
>>>>> ENDIF(COMMAND CMAKE_POLICY)
>>>>>
>>>>>
>>>>> # This project is designed to be built outside the Insight sourcetree.
>>>>> PROJECT(ScalarMultiPhase2DTest)
>>>>>
>>>>> # Find ITK.
>>>>> FIND_PACKAGE(ITK REQUIRED)
>>>>> INCLUDE(${ITK_USE_FILE})
>>>>>
>>>>> ADD_EXECUTABLE(ScalarMultiPhase2DTest ScalarMultiPhase2DTest.cxx )
>>>>>
>>>>> TARGET_LINK_LIBRARIES(ScalarMultiPhase2DTest ITKCommon)
>>>>> TARGET_LINK_LIBRARIES(ScalarMultiPhase2DTest ITKNumerics ITKIO)
>>>>> TARGET_LINK_LIBRARIES(ScalarMultiPhase2DTest ITKReview)
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Kishore Mosaliganti wrote:
>>>>>>
>>>>>> Hi Surcin,
>>>>>>
>>>>>> You don't have to make any changes in the ITK source folder. After
>>>>>> downloading ITK, use cmake and build it.
>>>>>>
>>>>>> In your project folder, you just need the main.cxx file with a
>>>>>> CMakeLists.txt and build it.
>>>>>>
>>>>>> You don't need the Insight Journal classes at all.
>>>>>>
>>>>>> Kishore
>>>>>>
>>>>>> On Tue, Jun 22, 2010 at 7:23 AM, julien_s <surchenjohn at gmail.com> wrote:
>>>>>>>
>>>>>>> First, thank you for your help
>>>>>>>
>>>>>>> Now,I compiled ITK with ITK_USE_REVIEW turned on.
>>>>>>>
>>>>>>> About the Review folder, if I understand, I have to replace the classes
>>>>>>> downloaded from the Insight Journal by the classes available in the
>>>>>>> Review
>>>>>>> Folder?
>>>>>>> I tried to do that but I have a lot of errors so I think I didn't
>>>>>>> understand.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Kishore Mosaliganti wrote:
>>>>>>>>
>>>>>>>> Hi Surcin,
>>>>>>>>
>>>>>>>> These set of filters were moved to the ITK Review folder and
>>>>>>>> subsequent changes were made to the filters directly in the Review
>>>>>>>> folders.
>>>>>>>>
>>>>>>>> Therefore, in your project, you should compile ITK with the
>>>>>>>> ITK_USE_REVIEW turned on in the cmake configuration. You can eliminate
>>>>>>>> the classes that you downloaded from the Insight Journal.
>>>>>>>>
>>>>>>>> Can you then repeat your example and let us know how it goes.
>>>>>>>>
>>>>>>>> Kishore
>>>>>>>>
>>>>>>>> On Mon, Jun 21, 2010 at 1:03 PM, Surcin Julien <juliensurcin at gmail.com>
>>>>>>>> wrote:
>>>>>>>>> Dear ITK users,
>>>>>>>>>
>>>>>>>>> I'm working on the Chan and Vese method of segmentation.
>>>>>>>>> Consequently I tried to use the example of the next publication:
>>>>>>>>>
>>>>>>>>> http://www.insight-journal.org/browse/publication/322
>>>>>>>>>
>>>>>>>>> I have no errors during the compilation but I have execution errors.
>>>>>>>>>
>>>>>>>>> "Debug Assertion Failed!
>>>>>>>>>
>>>>>>>>> Program:
>>>>>>>>> ..eContoursWhithoutEdges\
>>>>>>>>> Debug\ScalarSinglePhaseSparse2DTest.exe
>>>>>>>>> File:C:\Program Files\Microsoft Visual Studio 9.0\include\vector
>>>>>>>>> Line:779
>>>>>>>>>
>>>>>>>>> Expression:vector subscript out of range"
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> In fact, it is the same error that in the next post :
>>>>>>>>> http://www.itk.org/pipermail/insight-users/2009-May/030324.html
>>>>>>>>>
>>>>>>>>> (I use Visual Studio 2008 on Windows Vista)
>>>>>>>>>
>>>>>>>>> I'm searching a solution for one week but I don't find a solution.
>>>>>>>>>
>>>>>>>>> Thanks in advance,
>>>>>>>>>
>>>>>>>>> Julien Surcin
>>>>>>>>>
>>>>>>>>> Student in 4th year at Polytech'Orleans, School of Engineering, FRANCE
>>>>>>>>>
>>>>>>>>> _____________________________________
>>>>>>>>> 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
>>>>>>>>>
>>>>>>>>>
>>>>>>>> _____________________________________
>>>>>>>> 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
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> View this message in context:
>>>>>>> http://old.nabble.com/problem-with-Chan-and-Vese-method-of-segmentation-tp28951053p28958913.html
>>>>>>> Sent from the ITK - Users mailing list archive at Nabble.com.
>>>>>>>
>>>>>>> _____________________________________
>>>>>>> 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
>>>>>>>
>>>>>> _____________________________________
>>>>>> 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
>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> View this message in context: http://old.nabble.com/problem-with-Chan-and-Vese-method-of-segmentation-tp28951053p28968627.html
>>>>> Sent from the ITK - Users mailing list archive at Nabble.com.
>>>>>
>>>>> _____________________________________
>>>>> 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
>>>>>
>>>> _____________________________________
>>>> 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
>>>>
>>>
>>
>

More information about the Insight-users mailing list