From norman-k-williams@uiowa.edu Mon Jan 5 14:23:45 2004 From: norman-k-williams@uiowa.edu (kent williams) Date: Mon, 5 Jan 2004 08:23:45 -0600 Subject: [Insight-developers] Suggestion for ITK cvs tagging Message-ID: <006601c3d397$88cf3830$6a00a8c0@LAPCRUSHER> This is a multi-part message in MIME format. ------=_NextPart_000_0063_01C3D365.3C99B5D0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable At Iowa we're in the position of needing to build against the CVS = version of ITK for our development of Brains2, but don't necessarily = want to be at the bleeding edge of the daily updates. It would be nice if some non-branch tags were dropped on days when the = dashboard is all or mostly green, so that there were some intermediate = 'known-mostly-good' snapshots we could compile against. With developer = access I probably have priveleges to do top-level tags myself, but I = hesitate to do so -- not only does it thrash the CVS Repository, but it = engenders confusion if we get into a tagging free-for-all. There's also = the problem that CVS tagging isn't an atomic operation -- people do = checkins all the time when it's daytime in the US. ------=_NextPart_000_0063_01C3D365.3C99B5D0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

At Iowa we're in the position of = needing to build=20 against the CVS version of ITK for our development of Brains2, but don't = necessarily want to be at the bleeding edge of the daily = updates.

It would be nice if some non-branch = tags were=20 dropped on days when the dashboard is all or mostly green, so that there = were=20 some intermediate 'known-mostly-good' snapshots we could compile = against. =20 With developer access I probably have priveleges to do top-level tags = myself,=20 but I hesitate to do so -- not only does it thrash the CVS Repository, = but it=20 engenders confusion if we get into a tagging free-for-all. There's = also=20 the problem that CVS tagging isn't an atomic operation -- people do = checkins all=20 the time when it's daytime in the US.

------=_NextPart_000_0063_01C3D365.3C99B5D0-- From jjomier@cs.unc.edu Mon Jan 5 14:51:06 2004 From: jjomier@cs.unc.edu (Julien Jomier) Date: Mon, 5 Jan 2004 09:51:06 -0500 Subject: [Insight-developers] Suggestion for ITK cvs tagging References: <006601c3d397$88cf3830$6a00a8c0@LAPCRUSHER> Message-ID: <003d01c3d39b$59cf8630$13a81798@vangogh> This is a multi-part message in MIME format. ------=_NextPart_000_003A_01C3D371.70066BA0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Kent, How about using the date instead of a cvs tag?=20 We just need to know the date of the most recent green dashboard. I don't know if Dart can update an html page and sets this date = automatically (even different dates for different platforms). This will = be a nice thing to have. my two cents, Happy new year, Julien ----- Original Message -----=20 From: kent williams=20 To: insight developers=20 Sent: Monday, January 05, 2004 9:23 AM Subject: [Insight-developers] Suggestion for ITK cvs tagging At Iowa we're in the position of needing to build against the CVS = version of ITK for our development of Brains2, but don't necessarily = want to be at the bleeding edge of the daily updates. It would be nice if some non-branch tags were dropped on days when the = dashboard is all or mostly green, so that there were some intermediate = 'known-mostly-good' snapshots we could compile against. With developer = access I probably have priveleges to do top-level tags myself, but I = hesitate to do so -- not only does it thrash the CVS Repository, but it = engenders confusion if we get into a tagging free-for-all. There's also = the problem that CVS tagging isn't an atomic operation -- people do = checkins all the time when it's daytime in the US. ------=_NextPart_000_003A_01C3D371.70066BA0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Hi Kent,

How about using the date instead = of a cvs tag?=20

We just need to know the date of the = most recent=20 green dashboard.

I don't know if Dart can update an = html page=20 and sets this date automatically (even different dates for different = platforms).=20 This will be a nice thing to have.

my two cents,

Happy new year,

Julien

----- Original Message -----
From:=20 kent williams
To: insight = developers=20

Sent: Monday, January 05, 2004 = 9:23=20 AM

Subject: [Insight-developers] = Suggestion=20 for ITK cvs tagging

At Iowa we're in the position of = needing to build=20 against the CVS version of ITK for our development of Brains2, but = don't=20 necessarily want to be at the bleeding edge of the daily = updates.

It would be nice if some non-branch = tags were=20 dropped on days when the dashboard is all or mostly green, so that = there were=20 some intermediate 'known-mostly-good' snapshots we could compile=20 against. With developer access I probably have priveleges to do=20 top-level tags myself, but I hesitate to do so -- not only does it = thrash the=20 CVS Repository, but it engenders confusion if we get into a tagging=20 free-for-all. There's also the problem that CVS tagging isn't an = atomic=20 operation -- people do checkins all the time when it's daytime in the=20 US.

------=_NextPart_000_003A_01C3D371.70066BA0-- From norman-k-williams@uiowa.edu Mon Jan 5 22:23:11 2004 From: norman-k-williams@uiowa.edu (Kent Williams) Date: 05 Jan 2004 16:23:11 -0600 Subject: [Insight-developers] itk::array< itk::vector > - broken? In-Reply-To: <3FE48C3C.6000302@unc.edu> References: <3FE48C3C.6000302@unc.edu> Message-ID: <1073341390.25181.21.camel@nissl.psychiatry.uiowa.edu> I ran into this in a 'user-level' program, with GCC 3.3.2; is there a better way to fix this than to include the txx files in programs that use ITK libraries? On Sat, 2003-12-20 at 11:51, Stephen R. Aylward wrote: > Thanks! > > Adding > > #include "vnl/vnl_vector.txx" > #include "vnl/vnl_c_vector.txx" > > did the trick! > > Stephen > > Lydia Ng wrote: > > I think I had a similar problem before due to itk::Vector not being one of > > the standard types already instantiated in the vnl library - you might need > > to include vnl/vnl_vector.txx to get an itk::Array (a.k.a) vnl_vector of > > itk::Vector. > > > > - Lydia > > > > > >>-----Original Message----- > >>From: Stephen R. Aylward [mailto:aylward@unc.edu] > >>Sent: Friday, December 19, 2003 8:18 PM > >>To: insight-developers@public.kitware.com > >>Subject: [Insight-developers] itk::array< itk::vector > - broken? > >> > >> > >>Hi, > >> > >>I am getting odd errors under .net2003 using a checkout from today... > >> > >>A small sample: > >> > >>---- > >>#include "itkVector.h" > >>#include "itkArray.h" > >> > >>int main(int argc, char ** argv) > >> { > >> typedef itk::Vector VectorType; > >> typedef itk::Array< VectorType > ArrayType; > >> > >> ArrayType a; > >> a.SetSize(3); > >> > >> VectorType v; > >> v[0] = 1; > >> v[1] = 0; > >> v[2] = 0; > >> a[0] = v; > >> } > >>---- > >> > >>Produces the following two errors: > >> > >>test.obj : error LNK2019: unresolved external symbol "public: __thiscall > >>vnl_vector >::~vnl_vector >>itk::Vector >(void)" > >>(??1?$vnl_vector@V?$Vector@N$02@itk@@@@QAE@XZ) referenced in function > >>"public: __thiscall itk::Array > >> >::~Array >(void)" > >>(??1?$Array@V?$Vector@N$02@itk@@@itk@@QAE@XZ) > >> > >>test.obj : error LNK2019: unresolved external symbol "public: bool > >>__thiscall vnl_vector >::set_size(unsigned > >>int)" (?set_size@?$vnl_vector@V?$Vector@N$02@itk@@@@QAE_NI@Z) referenced > >>in function "public: void __thiscall itk::Array >>itk::Vector >::SetSize(unsigned int)" > >>(?SetSize@?$Array@V?$Vector@N$02@itk@@@itk@@QAEXI@Z) > >>C:\aylward\src\itkUNCApplications-VC++\bin\RelWithDebInfo\test.exe : > >>fatal error LNK1120: 2 unresolved externals > >> > >>---- > >>Seems like this should work - I thought I had done this in the past, but > >>can't find any reference code. Any ideas? > >> > >>The CMakeLists.txt file is > >>PROJECT(test) > >> > >>INCLUDE (${CMAKE_ROOT}/Modules/FindITK.cmake) > >>IF(USE_ITK_FILE) > >> INCLUDE(${USE_ITK_FILE}) > >>ENDIF(USE_ITK_FILE) > >> > >>LINK_DIRECTORIES ( > >> ${ITK_LIBRARY_PATH} > >>) > >> > >>LINK_LIBRARIES ( > >> ITKCommon > >>) > >> > >>SET ( MAIN_SRCS > >>test.cxx > >>) > >> > >>ADD_EXECUTABLE( test MAIN_SRCS ) > >>---- > >> > >>Thanks, > >>Stephen > >> > >>-- > >>=========================================================== > >>Dr. Stephen R. Aylward > >>Associate Professor of Radiology > >>Adjunct Associate Professor of Computer Science and Surgery > >>http://caddlab.rad.unc.edu > >>aylward@unc.edu > >>(919) 966-9695 > >> > >> > >>_______________________________________________ > >>Insight-developers mailing list > >>Insight-developers@itk.org > >>http://www.itk.org/mailman/listinfo/insight-developers > > > > > > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers From cates@sci.utah.edu Tue Jan 6 17:24:25 2004 From: cates@sci.utah.edu (Joshua Cates) Date: Tue, 6 Jan 2004 10:24:25 -0700 (MST) Subject: [Insight-developers] Re: IEEE Vis Tutorial Evaluations : ITK --> SPIE In-Reply-To: <3FFACAC8.6050001@kitware.com> Message-ID: Hi Luis, Sounds like a fine plan. The segmentation section notes from MICCAI are closest to what I will present at SPIE. thanks, Josh. ______________________________ Josh Cates Scientific Computing and Imaging Institute University of Utah Email: cates@sci.utah.edu Phone: (801) 587-7697 URL: http://www.sci.utah.edu/~cates On Tue, 6 Jan 2004, Luis Ibanez wrote: > > Here is the evaluation of the ITK course > presented at IEEE Visualization 2003. > > Based on the comments we may want to reconsider > adding the slides printouts as part of the notes > to be delivered to SPIE Medical Imaging. > > We are allowed to produce a 200 pages document. > Our current PDF text have only 100 pages. We could > use the remaining 100 for printing out slides at > 2 or 4 per page. > > If we do this for the PPT presentations in the > tutorial (at 4 slides per page ) we will have: > > Presentation # Pages > > Getting Started I 8 > Getting Started II 8 > Getting Started III 5 > Getting Started IV 11 > Getting Started V 13 > Registration (Lydia) 22 > Segmentation (Josh) 19 > HybridSegm.(Celina) 13 > > Total 99 > > > The deadline for submitting the notes is > January 9th (this friday). > > > > Any comments and suggestions are welcome, > > > > > Luis > > > ----------------------------------------------- > From ci42@columbia.edu Tue Jan 6 19:51:49 2004 From: ci42@columbia.edu (Celina Imielinska) Date: Tue, 6 Jan 2004 14:51:49 -0500 (EST) Subject: [Insight-developers] Re: IEEE Vis Tutorial Evaluations : ITK --> SPIE In-Reply-To: References: Message-ID: Luis, where should we send our presentations? with my 13 pages, am I allowed to have 52 slides? -Celina On Tue, 6 Jan 2004, Joshua Cates wrote: > Hi Luis, > > Sounds like a fine plan. The segmentation section notes from MICCAI are > closest to what I will present at SPIE. > > thanks, > > Josh. > > ______________________________ > Josh Cates > Scientific Computing and Imaging Institute > University of Utah > Email: cates@sci.utah.edu > Phone: (801) 587-7697 > URL: http://www.sci.utah.edu/~cates > > > On Tue, 6 Jan 2004, Luis Ibanez wrote: > > > > > Here is the evaluation of the ITK course > > presented at IEEE Visualization 2003. > > > > Based on the comments we may want to reconsider > > adding the slides printouts as part of the notes > > to be delivered to SPIE Medical Imaging. > > > > We are allowed to produce a 200 pages document. > > Our current PDF text have only 100 pages. We could > > use the remaining 100 for printing out slides at > > 2 or 4 per page. > > > > If we do this for the PPT presentations in the > > tutorial (at 4 slides per page ) we will have: > > > > Presentation # Pages > > > > Getting Started I 8 > > Getting Started II 8 > > Getting Started III 5 > > Getting Started IV 11 > > Getting Started V 13 > > Registration (Lydia) 22 > > Segmentation (Josh) 19 > > HybridSegm.(Celina) 13 > > > > Total 99 > > > > > > The deadline for submitting the notes is > > January 9th (this friday). > > > > > > > > Any comments and suggestions are welcome, > > > > > > > > > > Luis > > > > > > ----------------------------------------------- > > > From luis.ibanez@kitware.com Tue Jan 6 20:00:14 2004 From: luis.ibanez@kitware.com (Luis Ibanez) Date: Tue, 06 Jan 2004 15:00:14 -0500 Subject: [Insight-developers] Re: IEEE Vis Tutorial Evaluations : ITK --> SPIE In-Reply-To: References: Message-ID: <3FFB13CE.90308@kitware.com> Celina, I calculated 13 pages for your presentation based on the file ITK-SPIE2003-COLUMBIA.ppt http://www.itk.org/cgi-bin/cvsweb.cgi/InsightDocuments/CourseWare/Training/ITK-SPIE2003_COLUMBIA.ppt?cvsroot=Insight that was used for the 2003 SPIE tutorial. This file has 51 slides. If we printed out handouts at 4 slides/page it will result in 13 pages. Do you anticipate to make any changes to this presentation ? Please keep in mind that the deadline for submission is january 9th. Regards, Luis ------------------------ Celina Imielinska wrote: > Luis, > > where should we send our presentations? with my 13 pages, am I allowed > to have 52 slides? > > -Celina > > > On Tue, 6 Jan 2004, Joshua Cates wrote: > > >>Hi Luis, >> >>Sounds like a fine plan. The segmentation section notes from MICCAI are >>closest to what I will present at SPIE. >> >>thanks, >> >>Josh. >> >>______________________________ >> Josh Cates >> Scientific Computing and Imaging Institute >> University of Utah >> Email: cates@sci.utah.edu >> Phone: (801) 587-7697 >> URL: http://www.sci.utah.edu/~cates >> >> >>On Tue, 6 Jan 2004, Luis Ibanez wrote: >> >> >>>Here is the evaluation of the ITK course >>>presented at IEEE Visualization 2003. >>> >>>Based on the comments we may want to reconsider >>>adding the slides printouts as part of the notes >>>to be delivered to SPIE Medical Imaging. >>> >>>We are allowed to produce a 200 pages document. >>>Our current PDF text have only 100 pages. We could >>>use the remaining 100 for printing out slides at >>>2 or 4 per page. >>> >>>If we do this for the PPT presentations in the >>>tutorial (at 4 slides per page ) we will have: >>> >>> Presentation # Pages >>> >>>Getting Started I 8 >>>Getting Started II 8 >>>Getting Started III 5 >>>Getting Started IV 11 >>>Getting Started V 13 >>>Registration (Lydia) 22 >>>Segmentation (Josh) 19 >>>HybridSegm.(Celina) 13 >>> >>> Total 99 >>> >>> >>>The deadline for submitting the notes is >>>January 9th (this friday). >>> >>> >>> >>>Any comments and suggestions are welcome, >>> >>> >>> >>> >>> Luis >>> >>> >>>----------------------------------------------- >>> >> > From ci42@columbia.edu Tue Jan 6 20:33:52 2004 From: ci42@columbia.edu (Celina Imielinska) Date: Tue, 6 Jan 2004 15:33:52 -0500 (EST) Subject: [Insight-developers] Re: IEEE Vis Tutorial Evaluations : ITK --> SPIE In-Reply-To: <3FFB13CE.90308@kitware.com> References: <3FFB13CE.90308@kitware.com> Message-ID: I will make slight changes without adding new material... and will be ready by Friday... -Celina On Tue, 6 Jan 2004, Luis Ibanez wrote: > Celina, > > I calculated 13 pages for your presentation > based on the file > > ITK-SPIE2003-COLUMBIA.ppt > > http://www.itk.org/cgi-bin/cvsweb.cgi/InsightDocuments/CourseWare/Training/ITK-SPIE2003_COLUMBIA.ppt?cvsroot=Insight > > that was used for the 2003 SPIE tutorial. > > This file has 51 slides. If we printed > out handouts at 4 slides/page it will > result in 13 pages. > > Do you anticipate to make any changes > to this presentation ? > > Please keep in mind that the deadline > for submission is january 9th. > > > Regards, > > > Luis > > > ------------------------ > Celina Imielinska wrote: > > > Luis, > > > > where should we send our presentations? with my 13 pages, am I allowed > > to have 52 slides? > > > > -Celina > > > > > > On Tue, 6 Jan 2004, Joshua Cates wrote: > > > > > >>Hi Luis, > >> > >>Sounds like a fine plan. The segmentation section notes from MICCAI are > >>closest to what I will present at SPIE. > >> > >>thanks, > >> > >>Josh. > >> > >>______________________________ > >> Josh Cates > >> Scientific Computing and Imaging Institute > >> University of Utah > >> Email: cates@sci.utah.edu > >> Phone: (801) 587-7697 > >> URL: http://www.sci.utah.edu/~cates > >> > >> > >>On Tue, 6 Jan 2004, Luis Ibanez wrote: > >> > >> > >>>Here is the evaluation of the ITK course > >>>presented at IEEE Visualization 2003. > >>> > >>>Based on the comments we may want to reconsider > >>>adding the slides printouts as part of the notes > >>>to be delivered to SPIE Medical Imaging. > >>> > >>>We are allowed to produce a 200 pages document. > >>>Our current PDF text have only 100 pages. We could > >>>use the remaining 100 for printing out slides at > >>>2 or 4 per page. > >>> > >>>If we do this for the PPT presentations in the > >>>tutorial (at 4 slides per page ) we will have: > >>> > >>> Presentation # Pages > >>> > >>>Getting Started I 8 > >>>Getting Started II 8 > >>>Getting Started III 5 > >>>Getting Started IV 11 > >>>Getting Started V 13 > >>>Registration (Lydia) 22 > >>>Segmentation (Josh) 19 > >>>HybridSegm.(Celina) 13 > >>> > >>> Total 99 > >>> > >>> > >>>The deadline for submitting the notes is > >>>January 9th (this friday). > >>> > >>> > >>> > >>>Any comments and suggestions are welcome, > >>> > >>> > >>> > >>> > >>> Luis > >>> > >>> > >>>----------------------------------------------- > >>> > >> > > > > > From bahrahm@yahoo.com Wed Jan 7 17:39:47 2004 From: bahrahm@yahoo.com (Jisung Kim) Date: Wed, 7 Jan 2004 09:39:47 -0800 (PST) Subject: [Insight-developers] Farewell. Message-ID: <20040107173947.39790.qmail@web41313.mail.yahoo.com> Hello. All ITK developers. I am leaving the states next tuesday and going back home (south korea). I just want to say thank you allk the ITK developers. I learned alot and got lots of help from you guys. I am not sure what I am goint to do in Korea. I hope I can spread what I learn from ITK community in Korea. Luis, I owe you too much to list what I got from you. You were my mentor and friend. By the way, I finished the book you gave me. :) I wish you and your family the best. Jim (GE), your comments and suggestions makes my work better. However, I still don't fully understand the ITK pipeline. :) Bill (GE), thank you for keep me up to the other developers. I wouldn't forget the horrifying feeling when I got an error notice from you. I learned how a developer and leader should act. Will (Kitware), I asked several times to make modifications in CVS repository. You never complained about my silly mistakes. Thank you. You always encourage me to be a better member of ITK community. I wouldn't forget your kind words and appreciation even to the small achievmant I made. Lydia, I should say I copied alot of your code in my examples and applications. :) Thank you for teaching me many things. Sayan, I enjoyed alot working with you on ITK statistics. I really appreciate your support for the ITK statistics framework. Terry, thank you for making the ITK community so that I can learn and have fun with many good people. To me, ITK is more than a software library. Thank you Bill (Kitware), George, Jim (upenn) and all the developers. I don't think people like too long list of appreciation so I want to stop here. :) Sincerely, Jisung Kim. ===== Jisung Kim bahrahm@yahoo.com 106 Mason Farm Rd. 129 Radiology Research Lab., CB# 7515 Univ. of North Carolina at Chapel Hill Chapel Hill, NC 27599-7515 __________________________________ Do you Yahoo!? Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes http://hotjobs.sweepstakes.yahoo.com/signingbonus From norman-k-williams@uiowa.edu Wed Jan 7 21:55:09 2004 From: norman-k-williams@uiowa.edu (Kent Williams) Date: 07 Jan 2004 15:55:09 -0600 Subject: [Insight-developers] itk:FlipImageFilter Message-ID: <1073512509.25768.6.camel@nissl.psychiatry.uiowa.edu> Does the FlipImageFilter expect the argument to SetFlipAxes (a FlipAxesArrayType) to be arranged slowest-moving-dimension first, or last? From luis.ibanez@kitware.com Wed Jan 7 22:07:08 2004 From: luis.ibanez@kitware.com (Luis Ibanez) Date: Wed, 07 Jan 2004 17:07:08 -0500 Subject: [Insight-developers] itk:FlipImageFilter In-Reply-To: <1073512509.25768.6.camel@nissl.psychiatry.uiowa.edu> References: <1073512509.25768.6.camel@nissl.psychiatry.uiowa.edu> Message-ID: <3FFC830C.2070301@kitware.com> Kent, The array of FlipAxes follows the same order of ITK indices, which is that the lowest index corresponds to the fastest moving dimension. In other words Flip[0] --> Index[0] ---> contiguous pixels in memory Flip[1] --> Index[1] ---> pixels that are one row appart Flip[2] --> Index[2] ---> pixels that are one slice appart Flip[3] --> Index[3] ---> pixels that are one volume appart ....and so on Luis ---------------------- Kent Williams wrote: > Does the FlipImageFilter expect the argument to SetFlipAxes (a > FlipAxesArrayType) to be arranged slowest-moving-dimension first, or > last? > > > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers > From norman-k-williams@uiowa.edu Wed Jan 7 22:30:20 2004 From: norman-k-williams@uiowa.edu (Kent Williams) Date: 07 Jan 2004 16:30:20 -0600 Subject: [Insight-developers] itk:FlipImageFilter In-Reply-To: <3FFC830C.2070301@kitware.com> References: <1073512509.25768.6.camel@nissl.psychiatry.uiowa.edu> <3FFC830C.2070301@kitware.com> Message-ID: <1073514620.25768.24.camel@nissl.psychiatry.uiowa.edu> That is what I thought. Unfortunately that is not what I'm seeing -- In the OrienterImageFilter I am trying to flip the first dimension, so I'm passing in { true, false, false }. But when I dump out the pixel data, it appears it's reversing the slices, not the rows. On Wed, 2004-01-07 at 16:07, Luis Ibanez wrote: > Kent, > > The array of FlipAxes follows the same order of ITK indices, > which is that the lowest index corresponds to the fastest > moving dimension. > > > In other words > > Flip[0] --> Index[0] ---> contiguous pixels in memory > Flip[1] --> Index[1] ---> pixels that are one row appart > Flip[2] --> Index[2] ---> pixels that are one slice appart > Flip[3] --> Index[3] ---> pixels that are one volume appart > > ....and so on > > > Luis > > > ---------------------- > > Kent Williams wrote: > > > Does the FlipImageFilter expect the argument to SetFlipAxes (a > > FlipAxesArrayType) to be arranged slowest-moving-dimension first, or > > last? > > > > > > _______________________________________________ > > Insight-developers mailing list > > Insight-developers@itk.org > > http://www.itk.org/mailman/listinfo/insight-developers > > > > From luis.ibanez@kitware.com Wed Jan 7 23:46:22 2004 From: luis.ibanez@kitware.com (Luis Ibanez) Date: Wed, 07 Jan 2004 18:46:22 -0500 Subject: [Insight-developers] itk:FlipImageFilter In-Reply-To: <1073514620.25768.24.camel@nissl.psychiatry.uiowa.edu> References: <1073512509.25768.6.camel@nissl.psychiatry.uiowa.edu> <3FFC830C.2070301@kitware.com> <1073514620.25768.24.camel@nissl.psychiatry.uiowa.edu> Message-ID: <3FFC9A4E.5050103@kitware.com> Kent, Maybe what is in question is the meaning of "fliping". If you have an image formed by the following four characters: AB CD AB are along the X axis, thus the fastest moving axis = axis[0]. AC are along the Y axis, thus the slowest moving axis = axis[1]. If you process this image with the FlipImageFilter and provide a flip array = [1,0] the output image will look like BA DC If instead you provide a flip array = [0,1] the output image will look like CD AB A new example was added to the CVS repository under Insight/Examples/Filtering/FlipImageFilter.cxx The last two arguments in the command line are the parameters for the flip array. You may want to try this example and check if it corresponds to what you are observing. Luis ---------------------- Kent Williams wrote: > That is what I thought. Unfortunately that is not what I'm seeing -- > In the OrienterImageFilter I am trying to flip the first dimension, so > I'm passing in { true, false, false }. But when I dump out the pixel > data, it appears it's reversing the slices, not the rows. > > On Wed, 2004-01-07 at 16:07, Luis Ibanez wrote: > >>Kent, >> >>The array of FlipAxes follows the same order of ITK indices, >>which is that the lowest index corresponds to the fastest >>moving dimension. >> >> >>In other words >> >>Flip[0] --> Index[0] ---> contiguous pixels in memory >>Flip[1] --> Index[1] ---> pixels that are one row appart >>Flip[2] --> Index[2] ---> pixels that are one slice appart >>Flip[3] --> Index[3] ---> pixels that are one volume appart >> >>....and so on >> >> >>Luis >> >> >>---------------------- >> >>Kent Williams wrote: >> >> >>>Does the FlipImageFilter expect the argument to SetFlipAxes (a >>>FlipAxesArrayType) to be arranged slowest-moving-dimension first, or >>>last? >>> >>> >>>_______________________________________________ >>>Insight-developers mailing list >>>Insight-developers@itk.org >>>http://www.itk.org/mailman/listinfo/insight-developers >>> >> >> > > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers > From lng@insightful.com Thu Jan 8 13:07:50 2004 From: lng@insightful.com (Lydia Ng) Date: Thu, 8 Jan 2004 05:07:50 -0800 Subject: [Insight-developers] modifying BUILDNAME Message-ID: This is a multi-part message in MIME format. ------_=_NextPart_001_01C3D5E8.6B19BC10 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi, =20 Yesterday on one of my machine I did a new/clean ITK current cvs = download and updated CMake to the latest release. When I ran CMake via the Windows GUI, I seem to have lost the ability = to modify the BUILDNAME. e.g it uses the generic "Windows-devenv" for the dashboard - while I = wanted to mark two different experimental builds differently. Anyone else have a similar problem? =20 - Lydia ------_=_NextPart_001_01C3D5E8.6B19BC10 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hi,

Yesterday on one of my machine I did a = new/clean ITK current cvs download and updated CMake to the latest = release.

When I ran CMake via the Windows GUI, I seem to have lost the ability = to modify the BUILDNAME.

e.g it uses the generic = “Windows-devenv” for the dashboard – while I wanted to mark two different = experimental builds differently.

Anyone else have a similar = problem?

- Lydia

------_=_NextPart_001_01C3D5E8.6B19BC10-- From lorensen@crd.ge.com Thu Jan 8 13:44:57 2004 From: lorensen@crd.ge.com (Lorensen, William E (Research)) Date: Thu, 8 Jan 2004 08:44:57 -0500 Subject: [Insight-developers] modifying BUILDNAME Message-ID: This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C3D5ED.60E5D350 Content-Type: text/plain I suspect this is related to an unannounced change to cmake/ctest. If Dart is not found, then ctest is used as the client. I think ctest creates the build name auomatically. Make sure Dart is found, then reconfingure and see if that helps, Bill -----Original Message----- From: Lydia Ng [mailto:lng@insightful.com] Sent: Thursday, January 08, 2004 8:08 AM To: insight-developers@itk.org Subject: [Insight-developers] modifying BUILDNAME Hi, Yesterday on one of my machine I did a new/clean ITK current cvs download and updated CMake to the latest release. When I ran CMake via the Windows GUI, I seem to have lost the ability to modify the BUILDNAME. e.g it uses the generic "Windows-devenv" for the dashboard - while I wanted to mark two different experimental builds differently. Anyone else have a similar problem? - Lydia ------_=_NextPart_001_01C3D5ED.60E5D350 Content-Type: text/html Content-Transfer-Encoding: quoted-printable

I=20 suspect this is related to an unannounced change to cmake/ctest. If = Dart is not=20 found, then ctest is used as the client. I think ctest creates the = build name=20 auomatically. Make sure Dart is found, then reconfingure and see if = that=20 helps,

Bill

-----Original Message-----
From: Lydia Ng=20 [mailto:lng@insightful.com]
Sent: Thursday, January 08, = 2004 8:08=20 AM
To: insight-developers@itk.org
Subject:=20 [Insight-developers] modifying BUILDNAME

Hi,

Yesterday on one of my = machine I=20 did a new/clean ITK current cvs download and updated CMake to the = latest=20 release.

When I ran CMake via the = Windows=20 GUI, I seem to have = lost the=20 ability to modify the BUILDNAME.

e.g it uses the generic=20 “Windows-devenv” for the dashboard – while I wanted = to mark two different=20 experimental builds differently.

Anyone else have a = similar=20 problem?

-=20 Lydia

------_=_NextPart_001_01C3D5ED.60E5D350-- From millerjv@crd.ge.com Thu Jan 8 14:50:21 2004 From: millerjv@crd.ge.com (Miller, James V (Research)) Date: Thu, 8 Jan 2004 09:50:21 -0500 Subject: [Insight-developers] itk:FlipImageFilter Message-ID: As Luis pointed out, there are two possible conventions for specifying a flip. One is that you can "flip the data ALONG the specified axis". Here when you say to flip the x-axis, the flipped image would be the left-right mirror image of the input. As in Luis' example below, the data along this axis is reordered. Second is that you can "flip ABOUT the specified axis". In this case, when you say to flip the x-axis, the flipped image would be the up-down mirror image. FlipImageFilter uses the former convention, flip the data along the specified axis as opposed to flipping the data about the specified axis. -----Original Message----- From: Luis Ibanez [mailto:luis.ibanez@kitware.com] Sent: Wednesday, January 07, 2004 6:46 PM To: norman-k-williams@uiowa.edu Cc: insight-developers@itk.org Subject: Re: [Insight-developers] itk:FlipImageFilter Kent, Maybe what is in question is the meaning of "fliping". If you have an image formed by the following four characters: AB CD AB are along the X axis, thus the fastest moving axis = axis[0]. AC are along the Y axis, thus the slowest moving axis = axis[1]. If you process this image with the FlipImageFilter and provide a flip array = [1,0] the output image will look like BA DC If instead you provide a flip array = [0,1] the output image will look like CD AB A new example was added to the CVS repository under Insight/Examples/Filtering/FlipImageFilter.cxx The last two arguments in the command line are the parameters for the flip array. You may want to try this example and check if it corresponds to what you are observing. Luis ---------------------- Kent Williams wrote: > That is what I thought. Unfortunately that is not what I'm seeing -- > In the OrienterImageFilter I am trying to flip the first dimension, so > I'm passing in { true, false, false }. But when I dump out the pixel > data, it appears it's reversing the slices, not the rows. > > On Wed, 2004-01-07 at 16:07, Luis Ibanez wrote: > >>Kent, >> >>The array of FlipAxes follows the same order of ITK indices, >>which is that the lowest index corresponds to the fastest >>moving dimension. >> >> >>In other words >> >>Flip[0] --> Index[0] ---> contiguous pixels in memory >>Flip[1] --> Index[1] ---> pixels that are one row appart >>Flip[2] --> Index[2] ---> pixels that are one slice appart >>Flip[3] --> Index[3] ---> pixels that are one volume appart >> >>....and so on >> >> >>Luis >> >> >>---------------------- >> >>Kent Williams wrote: >> >> >>>Does the FlipImageFilter expect the argument to SetFlipAxes (a >>>FlipAxesArrayType) to be arranged slowest-moving-dimension first, or >>>last? >>> >>> >>>_______________________________________________ >>>Insight-developers mailing list >>>Insight-developers@itk.org >>>http://www.itk.org/mailman/listinfo/insight-developers >>> >> >> > > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers > _______________________________________________ Insight-developers mailing list Insight-developers@itk.org http://www.itk.org/mailman/listinfo/insight-developers From billlist@nycap.rr.com Thu Jan 8 15:06:43 2004 From: billlist@nycap.rr.com (William A. Hoffman) Date: Thu, 08 Jan 2004 10:06:43 -0500 Subject: [Insight-developers] modifying BUILDNAME In-Reply-To: References: Message-ID: <6.0.1.1.2.20040108095325.06b36a00@pop.nycap.rr.com> This is not a ctest/Dart issue. This problem comes from a change made in cmake 1.8. In cmake 1.6 BUILDNAME was a cache variable, and in 1.8 it was changed to be a non-cached variable. The BUILDNAME variable is still being used, but there is no way to change it with the CMakeSetup or ccmake. You can specify it on the command line with a -DBUILDNAME=, or you can manually edit the cache file, or pre-load the cache. In retrospect, I think this change was a mistake. So, at this point, to fix the problem, I propose the following: 1. Change cmake to cache the variable again. Put this change into 1.8.4. 2. Add the variable as a cache entry in the ITK CMakeLists.txt files, which means it will start working after an update to ITK, and we won't have to wait for cmake 1.8.4. 3. Lydia, if you want to change this right now, you can add the entry by hand into the cache, or use a -DBUILDNAME="name" on the command line to CMakeSetup or ccmake. If I don't hear any other ideas, I willl make the above changes today. Sorry for the confusion. -Bill At 08:44 AM 1/8/2004, Lorensen, William E (Research) wrote: >"urn:schemas-microsoft-com:office:office" xmlns:w = "urn:schemas-microsoft-com:office:word"> >I suspect this is related to an unannounced change to cmake/ctest. If Dart is not found, then ctest is used as the client. I think ctest creates the build name auomatically. Make sure Dart is found, then reconfingure and see if that helps, > >Bill >-----Original Message----- >From: Lydia Ng [mailto:lng@insightful.com] >Sent: Thursday, January 08, 2004 8:08 AM >To: insight-developers@itk.org >Subject: [Insight-developers] modifying BUILDNAME > >Hi, > > > >Yesterday on one of my machine I did a new/clean ITK current cvs download and updated CMake to the latest release. > >When I ran CMake via the Windows GUI, I seem to have lost the ability to modify the BUILDNAME. > >e.g it uses the generic �Windows-devenv� for the dashboard � while I wanted to mark two different experimental builds differently. > >Anyone else have a similar problem? > > > >- Lydia From millerjv@crd.ge.com Thu Jan 8 15:14:26 2004 From: millerjv@crd.ge.com (Miller, James V (Research)) Date: Thu, 8 Jan 2004 10:14:26 -0500 Subject: [Insight-developers] modifying BUILDNAME Message-ID: I like option 1 (and option 3 as a short term solution). Jim -----Original Message----- From: William A. Hoffman [mailto:billlist@nycap.rr.com] Sent: Thursday, January 08, 2004 10:07 AM To: Lorensen, William E (Research); 'Lydia Ng'; insight-developers@itk.org Subject: RE: [Insight-developers] modifying BUILDNAME This is not a ctest/Dart issue. This problem comes from a change made in cmake 1.8. In cmake 1.6 BUILDNAME was a cache variable, and in 1.8 it was changed to be a non-cached variable. The BUILDNAME variable is still being used, but there is no way to change it with the CMakeSetup or ccmake. You can specify it on the command line with a -DBUILDNAME=, or you can manually edit the cache file, or pre-load the cache. In retrospect, I think this change was a mistake. So, at this point, to fix the problem, I propose the following: 1. Change cmake to cache the variable again. Put this change into 1.8.4. 2. Add the variable as a cache entry in the ITK CMakeLists.txt files, which means it will start working after an update to ITK, and we won't have to wait for cmake 1.8.4. 3. Lydia, if you want to change this right now, you can add the entry by hand into the cache, or use a -DBUILDNAME="name" on the command line to CMakeSetup or ccmake. If I don't hear any other ideas, I willl make the above changes today. Sorry for the confusion. -Bill At 08:44 AM 1/8/2004, Lorensen, William E (Research) wrote: >"urn:schemas-microsoft-com:office:office" xmlns:w = "urn:schemas-microsoft-com:office:word"> >I suspect this is related to an unannounced change to cmake/ctest. If Dart is not found, then ctest is used as the client. I think ctest creates the build name auomatically. Make sure Dart is found, then reconfingure and see if that helps, > >Bill >-----Original Message----- >From: Lydia Ng [mailto:lng@insightful.com] >Sent: Thursday, January 08, 2004 8:08 AM >To: insight-developers@itk.org >Subject: [Insight-developers] modifying BUILDNAME > >Hi, > > > >Yesterday on one of my machine I did a new/clean ITK current cvs download and updated CMake to the latest release. > >When I ran CMake via the Windows GUI, I seem to have lost the ability to modify the BUILDNAME. > >e.g it uses the generic "Windows-devenv" for the dashboard - while I wanted to mark two different experimental builds differently. > >Anyone else have a similar problem? > > > >- Lydia _______________________________________________ Insight-developers mailing list Insight-developers@itk.org http://www.itk.org/mailman/listinfo/insight-developers From hjohnson@mail.psychiatry.uiowa.edu Fri Jan 9 01:49:28 2004 From: hjohnson@mail.psychiatry.uiowa.edu (Hans Johnson) Date: Thu, 08 Jan 2004 19:49:28 -0600 Subject: [Insight-developers] vcl compilation--Compiler flags not propogated correctly In-Reply-To: References: Message-ID: <3FFE08A8.5030506@mail.psychiatry.uiowa.edu> Brad, I spent some time today tracking this down. There were actually 2 separate problems. First is that the wrapping does not work on the icc compiler or the SGI compilers. Second, I had set custom compile flags, but not link flags, this is what now works: CMAKE_CXX_FLAGS:STRING= -g -UNDEBUG -w1 -wd810,188 -i_dynamic CMAKE_EXE_LINKER_FLAGS:STRING=' ' CMAKE_MODULE_LINKER_FLAGS:STRING= -i_dynamic -shared -fPIC -pthread CMAKE_SHARED_LINKER_FLAGS:STRING= -i_dynamic -shared -fPIC -pthread I know that if the CMAKE_EXE_LINKER_FLAGS is set teh same as the other, it also does not work. I will be trying to compile the wrapping periodically, because I really need that functionality. Thanks for all you help. Regards, Hans PS: I think you hard-coded the i_dynamic somewhere that can now be removed. I've seen compliations where this flag is set twice (no big deal, it will just cause problems at some point in the future if I needed to turn i_dynamic off). Brad King wrote: >Hans, > >Please remove this dashboard submission or change it to an Experimental >build until we resolve this problem. > >What version of redhat is the machine again? I'd like to find a machine >here with the same version to try to duplicate the problem. > >Thanks, >-Brad > > From brad.king@kitware.com Fri Jan 9 13:59:32 2004 From: brad.king@kitware.com (Brad King) Date: Fri, 9 Jan 2004 08:59:32 -0500 (EST) Subject: [Insight-developers] vcl compilation--Compiler flags not propogated correctly In-Reply-To: <3FFE08A8.5030506@mail.psychiatry.uiowa.edu> Message-ID: On Thu, 8 Jan 2004, Hans Johnson wrote: > First is that the wrapping does not work on the icc compiler or the SGI > compilers. What is the error message that you get? This is expected to be working. > Second, I had set custom compile flags, but not link flags, this is what > now works: > > CMAKE_CXX_FLAGS:STRING= -g -UNDEBUG -w1 -wd810,188 -i_dynamic > CMAKE_EXE_LINKER_FLAGS:STRING=' ' > CMAKE_MODULE_LINKER_FLAGS:STRING= -i_dynamic -shared -fPIC -pthread > CMAKE_SHARED_LINKER_FLAGS:STRING= -i_dynamic -shared -fPIC -pthread > > I know that if the CMAKE_EXE_LINKER_FLAGS is set teh same as the other, > it also does not work. So shared libraries have to be built with -i_dynamic and executables without it? > PS: I think you hard-coded the i_dynamic somewhere that can now be > removed. I've seen compliations where this flag is set twice (no big > deal, it will just cause problems at some point in the future if I > needed to turn i_dynamic off). We would like ITK to build on anyone's machine without manually adding these flags. I'd like to setup the listfiles to automatically add the needed flags when needed. Once we've decided what they are, I'll add the necessary code. -Brad From lorensen@crd.ge.com Fri Jan 9 14:06:27 2004 From: lorensen@crd.ge.com (Lorensen, William E (Research)) Date: Fri, 9 Jan 2004 09:06:27 -0500 Subject: [Insight-developers] T-Cons restarting today Message-ID: Folks, Just a reminder that we will have the first tcon of the year today. Same time, same place. Bill From luis.ibanez@kitware.com Fri Jan 9 14:39:33 2004 From: luis.ibanez@kitware.com (Luis Ibanez) Date: Fri, 09 Jan 2004 09:39:33 -0500 Subject: [Insight-developers] T-Cons restarting today In-Reply-To: References: Message-ID: <3FFEBD25.1010005@kitware.com> Can we do a Bug sorting in this Tcon ? We have a good number of new bugs logged during the holidays that may not have been proprely assigned or classified yet. Luis ----------------------- Lorensen, William E (Research) wrote: > Folks, > Just a reminder that we will have the first tcon of the year today. Same > time, same place. > > Bill > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers > From hans-johnson@uiowa.edu Fri Jan 9 19:27:47 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Fri, 09 Jan 2004 13:27:47 -0600 Subject: [Insight-developers] intel compiler CableSwig Problem-->CMake Message-ID: <1073676467.7109.25.camel@ipl-imac.psychiatry.uiowa.edu> Brad, Just a guess, but this may be a CMake problem. The include lines for the icc compiler always look in /opt/intel/compiler70/ia32/include/ first, and then in /usr/include/. Both of these directories have a sys subdirectory, but the /opt/intel/compiler70/ia32/include/sys only contains 2 files (not including stat.h). When I ran cmake, HAVE_SYS_STAT_H was not set. If I put a link from /usr/include/sys/stat.h into /opt/intel/ia32/include/sys/stat.h then cmake sets HAVE_SYS_STAT_H=1, then before compiling, I remove the link--> the result is a perfectly compiled library! It just looks just like what you said: we need CMake to recognize that I really do have HAVE_SYS_STAT_H=1. Thank you for all your help on this. Thanks, Hans From brad.king@kitware.com Fri Jan 9 20:01:38 2004 From: brad.king@kitware.com (Brad King) Date: Fri, 9 Jan 2004 15:01:38 -0500 (EST) Subject: [Insight-developers] Re: SGI Wrapping failing In-Reply-To: <1073678021.7113.29.camel@ipl-imac.psychiatry.uiowa.edu> Message-ID: On Fri, 9 Jan 2004, Hans J. Johnson wrote: > I've attached a log file that shows a build error on the SGI when > wrapping is turned on. This looks like another problem with way that > compiler functionality is automatically detected. There was a one character typo in the name of GCC-XML's support directory that makes it work with MIPSpro 7.3. I've committed a fix. -Brad From hans-johnson@uiowa.edu Sat Jan 10 16:50:22 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Sat, 10 Jan 2004 10:50:22 -0600 Subject: [Insight-developers] Errors in testing TCL Wrapping Message-ID: <1073753421.10518.6.camel@ipl-imac.psychiatry.uiowa.edu> Hello All, I'm trying to get a few dashboards working with TCL wrapping. I am not much of a TCL programmer, and really don't understand the intricacies of the wrapping mechanism. The following are the 3 tests that are failing (See dashboards for tesla from Jan 10, 2004). Standard Output: Standard Error: Error in startup script: invalid command name "itkStringStream" while executing "itkStringStream stream" (file "/scratch/purify/Testing/gcc3/Insight/Testing/Code/Common/itkBasicClasses.tcl" line 4) > > /scratch/purify/Testing/gcc3/DEBUG/Insight/Testing/Temporary/Build.log.21486 > /scratch/purify/Testing/gcc3/DEBUG/Insight/Testing/Temporary/Build.log.28383 > ^[^[:q > > > /scratch/purify/Testing/gcc3/DEBUG/Insight/Testing/Temporary/Build.log.21486 > /scratch/purify/Testing/gcc3/DEBUG/Insight/Testing/Temporary/Build.log.28383 > ^[^[:q > Please let me know if there is anything that needs to be changed in the test builds. Regards, Hans From millerjv@crd.ge.com Mon Jan 12 15:01:27 2004 From: millerjv@crd.ge.com (Miller, James V (Research)) Date: Mon, 12 Jan 2004 10:01:27 -0500 Subject: [Insight-developers] Hough transform examples Message-ID: This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C3D91C.F4385FA6 Content-Type: text/plain; charset="iso-8859-1" If you run the Hough transform examples with not command line arguments, they print out a message like: $ HoughTransform2DLinesImageFilter.exe Missing Parameters Usage: c:\projects\I2\Insight-net2003\bin\relwithdebinfo\HoughTransform2DLinesImage Filter.exe inputImage outputImage numberOfLines variance of the accumulator blurring (default = 5) radius of the disk to remove from the accumulator (default = 10) >From the last two lines, I infer the last two parameters do not have to be set. But if I do not specify them then the programs crash. The Hough transform filter may have useful defaults, but the examples are blindly passing in argv[4] and argv[5] regardless of whether the user provided them. I am not sure what the intentions were. Should the user of the examples not have to specify these arguments or do these values have to be specified by the user (in which case the printout about is merely a "suggestion" of reasonable values)? Jim Miller _____________________________________ Visualization & Computer Vision GE Research Bldg. KW, Room C218B P.O. Box 8, Schenectady NY 12301 millerjv@research.ge.com james.miller@research.ge.com (518) 387-4005, Dial Comm: 8*833-4005, Cell: (518) 505-7065, Fax: (518) 387-6981 ------_=_NextPart_001_01C3D91C.F4385FA6 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

If you run the = Hough transform=20 examples with not command line arguments, they print out a message = like:

$=20 HoughTransform2DLinesImageFilter.exe = = = = = =20

Missing Parameters =
Usage:=20 c:\projects\I2\Insight-net2003\bin\relwithdebinfo\HoughTransform2DLinesI= mageFilter.exe
inputImage=20
outputImage
numberOfLines
variance of the=20 accumulator blurring (default =3D 5)
radius of the disk to = remove from=20 the accumulator (default =3D 10)

From the last two = lines, I=20 infer the last two parameters do not have to be set. But if I do = not=20 specify them then the=20 programs crash. The Hough transform filter may have useful = defaults, but=20 the examples are blindly passing in argv[4] and argv[5] regardless of = whether=20 the user provided them.

I am not sure what = the=20 intentions were. Should the user of the examples not have to = specify these=20 arguments or do these values have to be specified by the user (in = which=20 case the printout about is merely a "suggestion" of reasonable=20 values)?

Jim = Miller=20
_____________________________________
Visualization &=20 Computer Vision
GE = Research
Bldg. KW, Room C218B
P.O. Box 8, = Schenectady NY=20 12301

millerjv@research.ge.com

james.miller@research.ge.com
(518) = 387-4005, Dial=20 Comm: 8*833-4005,
Cell: = (518) 505-7065,=20 Fax: (518) 387-6981

------_=_NextPart_001_01C3D91C.F4385FA6-- From jjomier@cs.unc.edu Mon Jan 12 15:11:20 2004 From: jjomier@cs.unc.edu (Julien Jomier) Date: Mon, 12 Jan 2004 10:11:20 -0500 Subject: [Insight-developers] Hough transform examples In-Reply-To: Message-ID: <003901c3d91e$55ec6d70$5d691398@pissarro> Hi Jim, That's a bug. The intention was to leave these values as options for the user since the example should work with the default values but can do a better job with optimal values. Let me know if I should fix this, or if you already did it, Thanks, Julien -----Original Message----- From: insight-developers-admin@itk.org [mailto:insight-developers-admin@itk.org] On Behalf Of Miller, James V (Research) Sent: Monday, January 12, 2004 10:01 AM To: Insight-developers (E-mail) Subject: [Insight-developers] Hough transform examples If you run the Hough transform examples with not command line arguments, they print out a message like: $ HoughTransform2DLinesImageFilter.exe Missing Parameters=20 Usage: c:\projects\I2\Insight-net2003\bin\relwithdebinfo\HoughTransform2DLinesIm= age Filter.exe inputImage=20 outputImage numberOfLines=20 variance of the accumulator blurring (default =3D 5)=20 radius of the disk to remove from the accumulator (default =3D 10)=20 >From the last two lines, I infer the last two parameters do not have to = be set. But if I do not specify them then the programs crash. The Hough transform filter may have useful defaults, but the examples are blindly passing in argv[4] and argv[5] regardless of whether the user provided = them. I am not sure what the intentions were. Should the user of the examples = not have to specify these arguments or do these values have to be specified = by the user (in which case the printout about is merely a "suggestion" of reasonable values)? Jim Miller=20 _____________________________________ Visualization & Computer Vision GE Research Bldg. KW, Room C218B P.O. Box 8, Schenectady NY 12301 millerjv@research.ge.com james.miller@research.ge.com (518) 387-4005, Dial Comm: 8*833-4005,=20 Cell: (518) 505-7065, Fax: (518) 387-6981=20 From hans-johnson@uiowa.edu Mon Jan 12 15:25:07 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Mon, 12 Jan 2004 09:25:07 -0600 Subject: [Insight-developers] TCL Wrapping Failure Message-ID: <1073921107.7056.4.camel@ipl-imac.psychiatry.uiowa.edu> Hello All, It looks like the TCL wrapping has a failure for all platforms except VC60. The Error is as follows: itkBasicClasses Failed Execution Time 0.460092 Completion Status Completed Exit Code CHILDSTATUS Exit Value 1 Test output Standard Output: Standard Error: Error in startup script: invalid command name "itkStringStream" while executing "itkStringStream stream" (file "/scratch/purify/Testing/gcc3/Insight/Testing/Code/Common/itkBasicClasses.tcl" line 4) There are also 2 other tests that are failing. Regards, Hans PS: I don't remember if I sent this to the list yesterday, but I did not want to forget it. If it was already submitted, please forgive me. From hans-johnson@uiowa.edu Mon Jan 12 15:41:27 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Mon, 12 Jan 2004 09:41:27 -0600 Subject: [Insight-developers] CMake Configuration Anomolies for redhat9 gcc 3.2 Message-ID: <1073922087.7055.13.camel@ipl-imac.psychiatry.uiowa.edu> Hello Build Experts, I've been trying to diagnose a problem that occurs when building with compiler optimizations in gcc 3.2 on redhat 9 system. The optimizations cause the Wrapping code to fail. I am fairly confident that it is because some of the automatically defined "HAVE_???" CMake variabls are being miss diagnosed. I must admit that I don't really understand how the automatic determination of compiler compatibility works. I have 2 builds (tesla.uiowa->Linux-g++-3.2-dbg, tesla.uiowa->Linux-g++-3.2-opt), the opt version fails. A completly clean build has been done for both of these. The main difference between the two builds is that one is "-g" and the other is "-O3". Here is the diff between the two CMakeCache.txt files: 17c17 < BUILDNAME:STRING=Linux-g++-3.2-dbg --- > BUILDNAME:STRING=Linux-g++-3.2-opt 74c74 < CMAKE_CXX_FLAGS:STRING= -g -fstrict-aliasing -UNDEBUG -Wall -Wcast-qual -Wstrict-prototypes --- > CMAKE_CXX_FLAGS:STRING= -DNDEBUG -O3 -funroll-loops -fno-omit-frame-pointer -fstrict-aliasing 95c95 < CMAKE_C_FLAGS:STRING= -g -UNDEBUG -Wall -Wstrict-prototypes --- > CMAKE_C_FLAGS:STRING= -DNDEBUG -O3 -funroll-loops -fno-omit-frame-pointer 1350c1350 < HAVE__DOPRNT_COMPILED:INTERNAL=FALSE --- > HAVE__DOPRNT_COMPILED:INTERNAL=TRUE 1352c1352 < HAVE__SYSTEM_CONFIGURATION_COMPILED:INTERNAL=FALSE --- > HAVE__SYSTEM_CONFIGURATION_COMPILED:INTERNAL=TRUE 1941,1942c1941,1942 < //Test < VCL_STATIC_CONST_INIT_NO_DEFN:INTERNAL=0 --- > //VXL test > VCL_STATIC_CONST_INIT_NO_DEFN:INTERNAL=1 I hope that this helps track down the problems. Regards, Hans From hans-johnson@uiowa.edu Mon Jan 12 16:10:26 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Mon, 12 Jan 2004 10:10:26 -0600 Subject: [Insight-developers] vnl test failing on SGI Message-ID: <1073923826.7055.24.camel@ipl-imac.psychiatry.uiowa.edu> Hello again, Would someone please forward this to the VNL folks if that is the appropriate space for it? There has been a single vnl test failure on SGI platform for some time now. The error occurs when trying to determine if a "long double" has a "nan" value. There is magic here that I don't understand. 141 vnl_math_isnan(long double x) { 142 if (sizeof(long double) == 8) return bMe(&x,0x7ff00000L,sz_l) && bMp(&x,0x000fffffL,sz_l); 143 else if (sizeof(long double) <= 12) return bMe(&x,0x4001ffffL,sz_l) && bMp(&x,0x40000000,sz_l-4); 144/*This is returning the wrong value*/ else return bMe(&x,0x7ff70000L,sz_l) && bMp(&x,0x0008ffffL,sz_l); 145 } The debugger reports that x has Value: nan0x000000010000000 But this function is returning false. My guess is that a long double on SGI is larger than that on all other platforms, and the the logic is wrong for when there are more than 12 bytes in a long double. Regards, Hans From brad.king@kitware.com Mon Jan 12 16:43:07 2004 From: brad.king@kitware.com (Brad King) Date: Mon, 12 Jan 2004 11:43:07 -0500 (EST) Subject: [Insight-developers] TCL Wrapping Failure In-Reply-To: <1073921107.7056.4.camel@ipl-imac.psychiatry.uiowa.edu> Message-ID: On Mon, 12 Jan 2004, Hans J. Johnson wrote: > It looks like the TCL wrapping has a failure for all platforms except > VC60. Actually, wrapping is failing on any platform with a CableSwig that has an upgraded GCC version. About a month ago I switched CableSwig to use GCC 3.3 instead of 3.0 for its internal parser. The XML dump changed the format of the list of strings used to initialize all the wrapper modules. Therefore CableSwig was getting an empty list of wrappers to load in the generated code. Since this change co-incided with the VXL upgrade in ITK, I did not notice the test failures until you pointed them out. I'm working not it now. Thanks, -Brad From brad.king@kitware.com Mon Jan 12 16:45:58 2004 From: brad.king@kitware.com (Brad King) Date: Mon, 12 Jan 2004 11:45:58 -0500 (EST) Subject: [Insight-developers] CMake Configuration Anomolies for redhat9 gcc 3.2 In-Reply-To: <1073922087.7055.13.camel@ipl-imac.psychiatry.uiowa.edu> Message-ID: On Mon, 12 Jan 2004, Hans J. Johnson wrote: > clean build has been done for both of these. The main difference > between the two builds is that one is "-g" and the other is "-O3". > Here is the diff between the two CMakeCache.txt files: > > 17c17 > < BUILDNAME:STRING=Linux-g++-3.2-dbg > --- > > BUILDNAME:STRING=Linux-g++-3.2-opt > 74c74 > < CMAKE_CXX_FLAGS:STRING= -g -fstrict-aliasing -UNDEBUG -Wall > -Wcast-qual -Wstrict-prototypes > --- > > CMAKE_CXX_FLAGS:STRING= -DNDEBUG -O3 -funroll-loops > -fno-omit-frame-pointer -fstrict-aliasing > 95c95 > < CMAKE_C_FLAGS:STRING= -g -UNDEBUG -Wall -Wstrict-prototypes > --- > > CMAKE_C_FLAGS:STRING= -DNDEBUG -O3 -funroll-loops > -fno-omit-frame-pointer > 1350c1350 > < HAVE__DOPRNT_COMPILED:INTERNAL=FALSE > --- > > HAVE__DOPRNT_COMPILED:INTERNAL=TRUE > 1352c1352 > < HAVE__SYSTEM_CONFIGURATION_COMPILED:INTERNAL=FALSE > --- > > HAVE__SYSTEM_CONFIGURATION_COMPILED:INTERNAL=TRUE > 1941,1942c1941,1942 > < //Test > < VCL_STATIC_CONST_INIT_NO_DEFN:INTERNAL=0 > --- > > //VXL test > > VCL_STATIC_CONST_INIT_NO_DEFN:INTERNAL=1 This is the same CMake bug that causes problems with the Intel compiler. Its built-in CHECK_VARIABLE_EXISTS test is always returning true when optimization is on. This was never noticed because CHECK_VARIABLE_EXISTS is rarely used and most dashboards use debug flags. This bug will be fixed in CMake 1.8.4. Until then, please just manually switch the values in the optimized build for your dashboard. Thanks, -Brad From brad.king@kitware.com Mon Jan 12 17:11:31 2004 From: brad.king@kitware.com (Brad King) Date: Mon, 12 Jan 2004 12:11:31 -0500 (EST) Subject: [Insight-developers] Upcoming ITK release Message-ID: Hello, As discussed in Friday's TCON, we need a release for SPIE within a few weeks. Since we've had the major vnl upgrade, I think making this the 1.6 release is justified. I need to fix several problems first (particularly with wrapping), and the dashboard needs to be cleaned up. I propose the following schedule to get this done ASAP: 1.) Freeze features at the end of today (Monday, Jan 12, 2004). 2.) Commit all fixes by Friday, freeze CVS at 5:00 EST Friday. 3.) Bill L. and I will do final cleanups over the weekend. 4.) Tag/branch ITK-1-6 on next Monday morning and re-open main CVS. Comments? -Brad From cates@sci.utah.edu Mon Jan 12 17:34:31 2004 From: cates@sci.utah.edu (Joshua Cates) Date: Mon, 12 Jan 2004 10:34:31 -0700 (MST) Subject: [Insight-developers] download stats Message-ID: Hi, Does anyone have the latest download stats and number of users on the mailing list for ITK? thanks, Josh. ______________________________ Josh Cates Scientific Computing and Imaging Institute University of Utah Email: cates@sci.utah.edu Phone: (801) 587-7697 URL: http://www.sci.utah.edu/~cates From norman-k-williams@uiowa.edu Mon Jan 12 22:27:33 2004 From: norman-k-williams@uiowa.edu (Kent Williams) Date: 12 Jan 2004 16:27:33 -0600 Subject: [Insight-developers] SpatialObject Spacing still double[3] Message-ID: <1073946453.25768.76.camel@nissl.psychiatry.uiowa.edu> This I noticed today, when I tried to compile some code that went something like this: itk::TubeSpatialObjectType tube; itk::ImageType image; tube->SetSpacing(image->GetSpacing()); This didn't work so hot because SpatialObject still defines SetSpacing this way: void SetSpacing( const double spacing[itkGetStaticConstMacro(ObjectDimension)] )]); Is this an oversight? From wlorens1@nycap.rr.com Tue Jan 13 03:09:27 2004 From: wlorens1@nycap.rr.com (Bill Lorensen) Date: Mon, 12 Jan 2004 22:09:27 -0500 Subject: [Insight-developers] Upcoming ITK release In-Reply-To: References: Message-ID: <6.0.0.22.2.20040112220911.02ddc008@pop.nycap.rr.com> Looks like a plan. Sounds good. At 12:11 PM 1/12/2004, Brad King wrote: >Hello, > >As discussed in Friday's TCON, we need a release for SPIE within a few >weeks. Since we've had the major vnl upgrade, I think making this the 1.6 >release is justified. I need to fix several problems first (particularly >with wrapping), and the dashboard needs to be cleaned up. > >I propose the following schedule to get this done ASAP: > >1.) Freeze features at the end of today (Monday, Jan 12, 2004). >2.) Commit all fixes by Friday, freeze CVS at 5:00 EST Friday. >3.) Bill L. and I will do final cleanups over the weekend. >4.) Tag/branch ITK-1-6 on next Monday morning and re-open main CVS. > >Comments? >-Brad > >_______________________________________________ >Insight-developers mailing list >Insight-developers@itk.org >http://www.itk.org/mailman/listinfo/insight-developers From wlorens1@nycap.rr.com Tue Jan 13 04:54:16 2004 From: wlorens1@nycap.rr.com (Bill Lorensen) Date: Mon, 12 Jan 2004 23:54:16 -0500 Subject: [Insight-developers] SpatialObject Spacing still double[3] In-Reply-To: <1073946453.25768.76.camel@nissl.psychiatry.uiowa.edu> References: <1073946453.25768.76.camel@nissl.psychiatry.uiowa.edu> Message-ID: <6.0.0.22.2.20040112232759.02d85b88@pop.nycap.rr.com> Yes it is an oversight. I guess we missed at least one. I'll take a look tomorrow. Bill At 05:27 PM 1/12/2004, Kent Williams wrote: >This I noticed today, when I tried to compile some code that went >something like this: > >itk::TubeSpatialObjectType tube; >itk::ImageType image; > >tube->SetSpacing(image->GetSpacing()); > >This didn't work so hot because SpatialObject still defines SetSpacing >this way: > >void SetSpacing( const double >spacing[itkGetStaticConstMacro(ObjectDimension)] )]); > >Is this an oversight? > > >_______________________________________________ >Insight-developers mailing list >Insight-developers@itk.org >http://www.itk.org/mailman/listinfo/insight-developers From will.schroeder@kitware.com Tue Jan 13 12:25:06 2004 From: will.schroeder@kitware.com (Will Schroeder) Date: Tue, 13 Jan 2004 07:25:06 -0500 Subject: [Insight-developers] download stats In-Reply-To: References: Message-ID: <6.0.1.1.2.20040113072407.03c5b328@pop.biz.rr.com> Hi Josh- I'll get that for you, the current stats page has some problems with disk space. Let me clean it up and get back. Will At 12:34 PM 1/12/2004, Joshua Cates wrote: >Hi, > >Does anyone have the latest download stats and number of users on the >mailing list for ITK? > >thanks, > >Josh. > >______________________________ > Josh Cates > Scientific Computing and Imaging Institute > University of Utah > Email: cates@sci.utah.edu > Phone: (801) 587-7697 > URL: http://www.sci.utah.edu/~cates > > > >_______________________________________________ >Insight-developers mailing list >Insight-developers@itk.org >http://www.itk.org/mailman/listinfo/insight-developers From millerjv@crd.ge.com Tue Jan 13 15:09:01 2004 From: millerjv@crd.ge.com (Miller, James V (Research)) Date: Tue, 13 Jan 2004 10:09:01 -0500 Subject: [Insight-developers] Hough transform examples Message-ID: This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C3D9E7.2BA7934C Content-Type: text/plain; charset="iso-8859-1" If you run the Hough transform examples with not command line arguments, they print out a message like: $ HoughTransform2DLinesImageFilter.exe Missing Parameters Usage: c:\projects\I2\Insight-net2003\bin\relwithdebinfo\HoughTransform2DLinesImage Filter.exe inputImage outputImage numberOfLines variance of the accumulator blurring (default = 5) radius of the disk to remove from the accumulator (default = 10) >From the last two lines, I infer the last two parameters do not have to be set. But if I do not specify them then the programs crash. The Hough transform filter may have useful defaults, but the examples are blindly passing in argv[4] and argv[5] regardless of whether the user provided them. I am not sure what the intentions were. Should the user of the examples not have to specify these arguments or do these values have to be specified by the user (in which case the printout about is merely a "suggestion" of reasonable values)? Jim Miller _____________________________________ Visualization & Computer Vision GE Research Bldg. KW, Room C218B P.O. Box 8, Schenectady NY 12301 millerjv@research.ge.com james.miller@research.ge.com (518) 387-4005, Dial Comm: 8*833-4005, Cell: (518) 505-7065, Fax: (518) 387-6981 ------_=_NextPart_001_01C3D9E7.2BA7934C Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

If you run the = Hough transform=20 examples with not command line arguments, they print out a message = like:

$=20 HoughTransform2DLinesImageFilter.exe = = = = = =20

james.miller@research.ge.com
(518) = 387-4005, Dial=20 Comm: 8*833-4005,
Cell: = (518) 505-7065,=20 Fax: (518) 387-6981

------_=_NextPart_001_01C3D9E7.2BA7934C-- From millerjv@crd.ge.com Wed Jan 14 14:24:00 2004 From: millerjv@crd.ge.com (Miller, James V (Research)) Date: Wed, 14 Jan 2004 09:24:00 -0500 Subject: [Insight-developers] Hough transform examples Message-ID: Julien, Can you handle this? Jim -----Original Message----- From: Julien Jomier [mailto:jjomier@cs.unc.edu] Sent: Monday, January 12, 2004 10:11 AM To: 'Miller, James V (Research)'; 'Insight-developers (E-mail)' Subject: RE: [Insight-developers] Hough transform examples Hi Jim, That's a bug. The intention was to leave these values as options for the user since the example should work with the default values but can do a better job with optimal values. Let me know if I should fix this, or if you already did it, Thanks, Julien -----Original Message----- From: insight-developers-admin@itk.org [mailto:insight-developers-admin@itk.org] On Behalf Of Miller, James V (Research) Sent: Monday, January 12, 2004 10:01 AM To: Insight-developers (E-mail) Subject: [Insight-developers] Hough transform examples If you run the Hough transform examples with not command line arguments, they print out a message like: $ HoughTransform2DLinesImageFilter.exe Missing Parameters Usage: c:\projects\I2\Insight-net2003\bin\relwithdebinfo\HoughTransform2DLinesImage Filter.exe inputImage outputImage numberOfLines variance of the accumulator blurring (default = 5) radius of the disk to remove from the accumulator (default = 10) >From the last two lines, I infer the last two parameters do not have to be set. But if I do not specify them then the programs crash. The Hough transform filter may have useful defaults, but the examples are blindly passing in argv[4] and argv[5] regardless of whether the user provided them. I am not sure what the intentions were. Should the user of the examples not have to specify these arguments or do these values have to be specified by the user (in which case the printout about is merely a "suggestion" of reasonable values)? Jim Miller _____________________________________ Visualization & Computer Vision GE Research Bldg. KW, Room C218B P.O. Box 8, Schenectady NY 12301 millerjv@research.ge.com james.miller@research.ge.com (518) 387-4005, Dial Comm: 8*833-4005, Cell: (518) 505-7065, Fax: (518) 387-6981 _______________________________________________ Insight-developers mailing list Insight-developers@itk.org http://www.itk.org/mailman/listinfo/insight-developers From samson@bwh.harvard.edu Wed Jan 14 14:55:11 2004 From: samson@bwh.harvard.edu (Samson Timoner) Date: Wed, 14 Jan 2004 09:55:11 -0500 Subject: [Insight-developers] correction Message-ID: Content-Type: multipart/mixed; boundary="++----------20040113203926-074758809----------++" This is a multi-part message in MIME format. --++----------20040113203926-074758809----------++ Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit When I submitted itkKLCompareHistogramImageToImageMetric, its name was changed to itkKullbackLieblerCompareHistogramImageToImageMetric. Alas, "Liebler" is a mis-spelling of "Leibler". I noticed this last week, and we should probably fix it before the next release. All the changes are in files that are attached. I submitted an experimental build from horus0.ai.mit.edu with the new files. The following files should be removed from CVS: Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetric.h Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetric.txx Testing/Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetricTes t.cxx The following files should be added to CVS: Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetric.h Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetric.txx Testing/Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetricTes t.cxx The following files should be changed, and are attached: Testing/Code/Algorithms/CMakeLists.txt Testing/Code/Algorithms/itkAlgorithmsHeaderTest.cxx Testing/Code/Algorithms/itkAlgorithmsTests2.cxx Testing/Code/Algorithms/itkCompareHistogramImageToImageMetricTest.cxx Note that I also made the following changes: 1) I added some comments to itkKullbackLeiblerCompareHistogramImageToImageMetr ic. 2) While my implementation of the algorithm was OK, it in fact did not correspond exactly to the paper I referenced. I therefore changed the code to make it do so. This changed 2 lines of code in itkKullbackLeiblerCompareHistogramImageToImageMetric.txx 3) I changed itkKullbackLeiblerCompareHistogramImageToImageMetricTest.cxx so that it would have a minimum at the appropriate location using the changed algorithm. I'm sorry to have caused problems, -- Samson P.S. If you do not like attachments, I put all the files in: http://www.ai.mit.edu/people/samson/upload/itk/ --++----------20040113203926-074758809----------++ Content-Type: application/x-gunzip; name="submit.tar.gz" Content-Transfer-Encoding: base64 H4sICGiVBEAAA3N1Ym1pdC50YXIA7F19c9s20s/f+hS4tHMjp471YjtOnbQzsl5sTSTbI8lxO70b DUxBFi4UqSMpvzSX734L8J0EQJB2097zRDOxZXJ/u4vFYrEAlkx3jD+REXU9d8978F78IZ9mq9l8 8+bgRbPZOjg6OoLf8Gnt+7+bB/sHzfaLFrvUarUPWi2gb785OHyBmn+MOunP1vWwg9ALF69d21LR Eedr6POVP8NBfTj7MD8Znncmv857w8lODSG4OOuO5vBvesb+RqjT681n/emsfulQy5sSc/m6Y97a DvVWaxd9/zki/8LJo8/3nxn36cXVpNtn3L80rjxqUo8St9HDjteI+HVXxPi05xkmQkUsuvaCNGLx TMP+eS+ttH8h27TatD+rd0anF5Ph7Gw85U2agoTuL7/w7/PLzuzsS4N6n2L2MwKDY0eIbOtA2zns Wb/T60+0hJ8RvCAO4wPKR52QIuEWZBTAKqfgFySm3Sli1i7DrV3Ibr8Mu/0MO4u6tufYG2oM7K3j rS4csMmI3BFzSrzhGt+SATU930wCOb6YMkyy8j3aMSl2T6iFnUdtgSpURsLWs9fYo8bM3timffs4 Ju5qCmoaRClCDUvL8NUYU2uMH7pb5w57W4cMTPtesz3aeKFY7H7a7+m1SoHIsrZPiVXAKyYRgs+J d287nzR4JCnTrLrYsh6n5HZNLA96w7ZKuqYuPiPVdhxicvquTZZLalDAn8FMbt86eM2xM5v/GhPP ASdRqVCFWVaf9QY7pLwC7UADHXhGJLghcctaWwVK8++RtW25E3ILSjncOMXsVZg09/6DRywXCD4S 0zb4lKjZhGJkTpKDDY8Npq5tWcTwyIJpCIoWiFHB0jIG2PW6K7xeEqcHTceWQTRbU4zMSxpjx1hR 67aIb5JOziWyZwmNi7AZaf1xOTeSAtJ8TwmkQi41OoZH7wj0kgfBsuSAKMNDQ/p0hTcEpnL7WfQo 5PYcGs1Fyc4TdJpn8qFTBy9YPP0IQ8h2Ssy5hcC0nCldb0wy2P7++2M0YC3iulMDm9jRlFmKSVp+ FLzH2OODQ1NkEU4ipdwUV2pm8W/abEZWhsgsnYDLZbfDHWYMnmT2XY9CxmYrzVEAE8hIhgpo0spe SPkjVABC2vznLWULZCB9/tJhqQTp8xctSgr57+vzP6jC/0Cf/2EV/of6/N9U4f9Gn/9RFf5H+vzf VuH/Vp//j1X4/1hifDUrDbBmCQnVhnCJMdyqNIhbJUZxq9IwbpUYx61KA7lVYiS3Kg3lVomx3Ko0 mFvZ0WzBlLyxYaWoXHZnyDI8XDtIpEouFAqBOTlP3TQqwSIt+wM219gaUYtgR2vOlwIyfNcEW65+ KiGjz3DdmuYNNj6NCL0xifPElXwlbmmFRnhjYoNi6wm7KmV4ZKQHROeE3q5ubGdl24tgLay2diFQ LCdYxT9WllfIIC0XsmyPuOOtt8Xm0Frazprbplw6XYpJRj745PTfW3AKV1eYFCHlXHGVoM8hI7ny vmoxMiNpMtDlnKPMchqeUOwOKDEXfAPQ0NubKMJlpGxNj06Ia5tbRnb5CBalC90m6ICz8jIuWdUT SvJJazEhxtZx6R15UvPLcVHbvdfvEd4WCMrJ2b6M+VU8lNLLr1Jr5ZigquIli9iyCgCb6iqIU+TS KrTzKmRcuILXyxz8HDuOfX+CrUWlXXhNuEzmbAUBemWbiyckCNW4ZTQi90oR/u0MhpnWpL+TReLU RTMo6WA1pF3a/NDd0w6G5bhIJ+XScnWwsvY+10RQnWNaswvP3Uaexim72DS2ZlEOX4TLSIF1in1r W9ic3mNv1e5dws/iwVAAy/coTE1049gGAHj/9OhySRwSLsoKzIrKckkrkPGF7FQki+k6yOw0zo67 TiFOtHvqiTpNl+VCLepR7kQlg5QGNHMGwDaNe8Tzk7KS0rTAAnn8MGTcuezarjfYWkZRXqEEyfiL gnM5YUUcMpI3mBn+4uZfYA6Jv0hlamAz0rYb4txRlwRD3MSuS5e0oL9UoAx/Nqy7K0wtVj2lFxRU mDT355iPq8/Cp/TmRnlMExFkz/fDZFYFTlOlObR7ejyydFlPW0HQ06uOEdGmuV1Z7go7G1ZAU7IH ipE7KCXqo2UOBjPlblBMslMbDupX0/4c/r7eCWv8UqfkcKOIX4ImLPKLeYYSpt3pSCiB3SiSkKDZ CWsQE5J83ulivPEwdVivDkq+FDkk3ZvB0W/uOLa4SwP76+EzQm3Htmzao5glN9KpL5AgIBayS45r bfULoWlR1xguuyuiscL2BcgAsoQj/KYxE7SzKYcMy9yJlYjO+uNLv9yVXaXWbaJ29UtjRtYb28HO Y74IKto5ko1uPhYkxEwqrzrN3dy78z6FNbOpmtb5dNKdstVpqarKPePhwQepKiMjKnVxY0imW5aY oRdUF8YUUc1g5lKiBDC8o1u7F9Fr7YbH1BXK8iJwrjuDG6oStZCmuMQsQamqEgvJstVYouuyWqqI VlYUFRKUKWVSYgpLfJ6GnrcjfFGFT0hYqiwnBBUV1uTo1B7lX46rXcLrlU58ooYJEpqUQHklTIpM vCmmQTJv6RC1dYj2dYgOdIgOdYje6BAd6RC91SH6UcuYTS0qLZu3tIze0rJ6S8vsLS27t2LDpw+6 o8vlJ0XpAXBEIDnKDe+XOe+MMEUHllnCwpPGEFDqeDACSY/5BBSa4UU/M8gfl8V31AdeEZ3OKU1I XO5sRyZCdSIjwWhEyzLnHhUw7RhTbkNVisuTBFvv0Z8aG+dKWun2s8A1C2mrbymHHIo2hSM69b5u UnntvdgQVLiXGrt6enc0vl64uRllCTo7kyli2fZinqhoWzDOVIo39SJa1c5cRKTYXwtpyuyKhZji fZwo/Qw3xuK1QXKbKrya3b5KzHPqiqxklq9+xiO5apEeB0bjR+/0ME9e2prC5W87Xv/mH16M1q+C xyQV96IkRi97fnLaLd9/yg9v2b5FSKm5x5Qmr+LYoq2mzL1L7HiUsZKvdor3lEJK2eZQ5Nb+NmHg KWxXpv9Lv3s165yM+ukHYePneJHkOmOyU5t1Jqf92Xw0PP8AP04mncmwP5WyGs4+JJ7Chr+GF5mH pws/ADrBsID1G8eZwDJb9jxvQnZqlyn5RPOX4EllmSkY2kX5S8wAOtqD4Pxm1Bct0/miJTJKGFNs NPZryvwJ0nPD1bFCW2AGHiRyjWyXbWVb2B49xUX79Ehf8HLp3e/7P5eae/QluLuGi3by+/HBw/69 zqwzn1xczDKPyxdHB9GZcZttnL5+bfiRNXgTQSTiSwNsSUxYsiVeRtBIShLK2dtYtym/8tkygfOL q9nl1czf79VlpNU6gcBEO4bWZus13OXBs+qGWqU7QfzWgefvhPZzNbT9F+4ErhtqZjuh9Oz7XH1S ftrXNURVzpXModOLH6nLcrJre42t/iOZmhQSXn+tE3KVNE6L1R9rmPSsd7jXRIfN1CV2XhW/XyUZ csMw3J1eD0/Zq1hYhFeeVs0Ms+2rfz2cniWmy9SLZFKHYopXwQiPsZgI9mKZ5EthEgr+2W/g+XM/ imz02WSo3//UOmrCvfT7nw4P9pvf3v/0NT6NVz8916cGI/bS4WvPYxizQ8ultysPJaMO+jtKLiLR zLbNT9RDbEyyUDG2F1uTHMO3EbZutxCagFP3hx/gQg/WYuzGR8hbAXoMwaZrbx4dLqNu7KA2eBA6 53yxiUb0hp2VI3uJxmRBDZiMah3TRJzeRRARiXNHFnuwvicEdS8uf50MT89m7CVoaGk7yIh4L4iH qekC5bNZ6lWj9h1dAuclqLWoz8fT7vxjf7JT+27j4Ns1RvfYYRM6qqMFddmuCzpGB0dv30Do/Y5Y C7qsAQPLMLcLgt5TG+xJ8PrnWqOBZivqoiUFxAq76IYQC90Si4C9yQLdPKKTLTUXiXEOcZHD2Dzv 2ShkSrCxQitO5nNjNgl6NCH7ZZkaALDtw8ssVlYKICKWVwQIqE+wadq2NbDhtpSluBpATgiehM1z mHP8DRU5Yf7UIdxiKYNR2yOoSthbia8H1Qq529KSBYWGOmUOIli07ckSQhFB4Z6VEJQuM2AdJ1fB dkgH/CbgfMUszbf13Y1tLdju9hiDnAcIXAZxXTGHckUYIh6anlCi99M7nwU8hVukAtp4m5cZdb+n IBTXkOiRypVV150IAIP+eMBOERLHFSH7kY0XYoSW3oPBjLmnSR7A4f61vWVnvPaEYLNAocGMEc3s LLgQdl1R3HUlecodeQlAVahTAFFTKgt3FO6iW/CjC81X7pQVrqz9ETEJCoAyp21FAUVZNySgLxeu VHVDAvLMEZck1nOqU7wFKmyl6xhk1KKaBxltmkqlQ1DBFB+YygjFlUcy6vxht4xSwFSlb1FvBUSp 48lSxHqaq876xPQVk8Khx9JVNqRM24W8lB8/SRv0YTSOz5fVjIU1Nrn8qPSxmkCSshhHEVJ0i3hE UFUtj4JeWdIjwIkPDkWEk87UxDdDSLI8HheENAN1q1SlNyJy3bIjEba4BuJpLIRlIRIuPlbX0lpV UWpgWXr9xoRLpktMnXTmLc+5Y9TMoZDVeNrAUguv0ksuvSKnYqC0bksDmqsPE2L0y4p04Fqo825X 02E16jeUqLL0ujvgIibSwrFCc2iXkSnBT+xKRQGYiFpRViYgl7qxVIAcEXpLdjrWtZiiGk06KvMl LhrsRdUwUgmT0xNeY8VXBFKrTPBjF9ZKiWpi3ynlI0i/elQI1suXCurzBAh2GF9lFctwVVax5cez uJBQYenCykMZRlSAWIZWsiTQq1ssDSxokOTpkwRKpq4ECQOjmsjcAy8ibOaZEhGJ52wN7wy7K2nk kZVnCrgpJxhFnzxhYrrS3JBWF4MKABrVfFJUVTuUO6gWMeCPGVcJPT6ySvDJViNqkJz6RzHCyVRd PyIHSOoYSwMkA1gAKhy7Ud3kib21FuCnOjR8ErtT8+v/e0vvsMnWABPIqG6IqabXVBPWAfSSbnjJ Cj/BBHcHX1hD6piLChFIS4HAbDOW7ujQOUSXTOVKWfq8jneWOV8uvfk+b2ANMhq0hkCH6oh93UUG xIxXCO3UPtf4/8fjEFggWaj5rval9v+8YuL/1oeKC1CfVYa6/qPZbh00M/UfB4eHrW/1H1/j02gg LyoZ8AsSXLhC+P98kygV5gUCrAiA3fPYHwsHlh8OKx6AMImwaQIEkYcNTOUuoowJ9tCjvUUrfEcQ Rstg+keQwJlkERSC+GUIbi0dmxosPkGPwGycC8CMfgyxCm6gWq12Z9MMr7oftWIWc1btQRb1t0dH h82j5s47uDvpnw6ns6BmXfYfaiFt0nYJ2n0ZreaLd2Xwaq/xFfNSvNFEDNB4Q4YYWKEeVcWo+E0p ErTolSdKUtXrTJRARYWwBCh9q4lETvzOnXe1ZKFRsqo/h0q+WeddWGaURYdV+zl08q05EfpbrvCX /2iHnSdkBWz+h9Arq/88aO5n6j/brWb74Nv8/zU+f4n6z7jykxd7T7pTlo4c6/+/cMw5d+/Q99LC 0aCM3P/ebjb3G612o/0jar05Pnh73GxxaFhaynUgd9T/o7XH7sLtdLlp1Dx76d2zhyBg0nFtiOzb 9R4SVpoixEpNh7MPEaM9Vm0KOdXK8zbHjcb9/f0eNHnPdm4bZ7PxqBETrrw1z77iYlTeIF7t6YYa wPcFsy292bJqz+vh7OziaoY657+i685k0jmf/foO3UMyYm8hU7sjVlDpz9M9SJQogICRgy3PL5zt T7pngOqcDEfD2a9M08Fwdt6fTtHgYoI66LIzmQ27V6POJOB0eTW5vJj29/ymMr74xobsL66mtWyP GpBissasbaZ0fLLxFyuxTRfcBIUrfgGHv6uYzUz5vdxVv95Aurf/9PKDXIIMHej6x038K7MvX+U3 Xr3yPYbn7xDQ4Nu9jdq912HrXFRntR0L7CzAdHfUH6X+ecGU/k4a7Z29WuR4nA17BgjczV80rPlh VbJP0Q3x7lkVMl83gDTKNAcHhg5iWw3aQ7ye3Jj47Z9+ju85j+gzVwiWIV2HgHW5EJdvHkeyGIEB wxMazUYtOATf5uASenTt17Khn1D7HSf1HjcEHAct7C3zi0v6QMwZXBPe7dq2s6AWSJ6QDRvsQaiL AYF+EBNsgynIxbopXmCF42N+/X0kbTet3s8QIu/CjDGvjRaHAdxZZBiE1usRw2RBJHQF5Du5QE/B UHif0S2lbPBeJ7HKImZpNZNqy1gJpR0f86UIcfK3gxtpJiIxMY/s3RSL0IjnNptfmKsvbejuexZg Ftyw3CdcxCxs8Eo5WAUjFmsffQtHjmnagFnGwmDU/fZP8M3PkCHtIgQ/v7wTINaJJkoh/MfStGFh npCwwQZAAcEhe83lLmI/A4xPnmQf0YvIhSIuIPZTy0c0OaLJEUgsQU3OfxR0+Dp7G/0UPE4mAZ6T +/qO3wK1GyQN53OOWIuBCc78R0611z/DEpf1WT3ThwFKDPBtVM9ZTQkKuq6e700lLC5GqC+x6aoV Y+dzpN4+ZF2WbHfWbnGz084eYITkQaOzjqWChE3O+bsKJGmwWICkvQLrXG0WEP/BG1io4A+m+KGC r+fZ0zfO1pIIiqFF2OzwkA4MGF0CHU/BylsPZvT8cJAOBBY48irnOIVBEoyGths/X/AzmPy0UDoL el/yvRfRJ92+3apsMhavyib8wATqZz75qS66fnwcpXfsTxT9xU7y1EDutC5HxV/T5Dlu0PUwjUGP sd1ghkz/mejjpKTI8fg1cJTkzWxsTGVn1gmFCROSssM373JsI08I2gNjlesEaSkPKj4i/Ou3JptL OMPMjVbmhq8mH9iRCB6kQgSoG/rxpbm9DdZOLKFjSts5r04wjD0tEY12cnITnpQM1AnBwQDCkP6G 6X1+BIlWAe9VyWro+tmkMeVkic5KXY87OtIJDJshSXR3or0RUT1uDsSOgF99Rxw9IEGl0ZLKdgQG UK+7shlr2Ho/q/8ZDRPMMw3P3orbntQImp8nTFggScvtwB/3z3W7yBQJ4yVF1FMGUZiwxxfF3FUd vrRDsEZ30P2KGquEA6N7apqwfguXeAuFR/tLxIRf8/h/suW1z8GDelyP7CQQecsmiia+mEywiW/H bsJlnG/XN8S5WMb0sRy2yVBPBZVP0C3Nd/DrfcbP4doPP4Tvo4mF/faJBQg2t6dVd3nUDLZkHHrH n1dgJ1u8WJLvZTDqRJz1Ec+lfDGbVIN84X5jWgH/RD/2oiZMPbIZEevWW/mq131kouPO7XvkxjM4 jF7K91GmHnS2YGzzu5pjXIDUDJp5cm2PlHCSh2EJQHsohpYcRsWV0lkjJqknJx52kojYJl4wUO+w uSX8GDZ2xr3cnDkm2N06/jy59r+zHDAQBUp+ZGzqsfPv5Ofd2FM4m1hc2iynSZ9KcNxNIMIGud7i +NhgrXn/Hr0cJ1v0E3rJLobKwldOTKyFmUmwGDSWGAITQzOJBV1DU/YfiGNQl/g2re+wFrBK3lRr /HuRnmLF+XncBlb2ECdfZsUh9jYTgz1Mh+p8cuo/GGTD4oT/BNbfEXlgY/VzgjFxHM44IgUG29uV 97c89wyEPAgoggqe/i/D2XzQGY6uJn1fsVr63vSq2+1Pp6zQ588+m/j2+eM/1Xacy8koeP/L4WE7 rv+Bv14025Dzfzv/+yqfv/r5XwXnLHUQeNBothqtfdRuHR8cHDePlAeB7W8Hgf+jB4EWWxbO51Uc ar6qfecfI1ZnkD6g0zvTq1mQNLkbbPBaPH7cxQ7w/sGPD1BpLfyOeYXQP24cCsboJs/tPrBFr4NG lNyY2FjVP4x2wtwyeX63CtmHvMA30ud6bCfyxl9Wsno88CNeHMjWK/4yIWLhp1ivQk7cgf2m8ZNF 8EL+yiK+MuUyWL4ZyONJPefsZ+Yhk5QOwcsJF3JBwbrWCWwweh02+TEAM4VT7XazDfdfSRcTMHde +29aYveZ9ibeMHKW/0OkYG80ge8hn3j7aJepLTQUGnrMJjfYZfbgPRHC41OmDd4Q5zjd1I4Jq0IP dfeme6i72lq3u+galvQU1BzvoWtimi4aDoe7QEgewJpg6XPbATmrXWbckM31Huozs4xgNJ86lE2T u4g/d/Z6bS+w6a+k04H95jFEs9r6NURQ0DH02deB06LwTQK7EET913RF3Hz3TPQt6+3AZ53XHddl /rXwt1ju2BPmIPQ1Gg+73c6QxfX2Lhqdd6eoffD2x92Qx2azhw7bh0B4uN/O+UW07wHGvqUsKkIv fhjVL+esm//zH/jNOmcnGkn/be9qm9u4jfD3+xWIPniOFkWKkpWkeklHcRxHrS17RLrTad1qKPJI XUPymLtjZMaj/vbuLl4OwOFeSCptx3PoNKYOwAJY7AG7i8VzF6y/mt9+Dnvt8OiR8WI+/dVis2jK fONRV9QXf7YklQe8qyRq0yhFOTorXMbR3fAunHG54b2ar2HxlAcUkkomojR3LOzxf46IDImZEMA1 u0PHami90hY7PoL447J++fbV4NVNX4351TIJZ8Br4NFwPIYZAKmFHSQmqkj0Lkrvtd50+Au3jHGb SdX7Mw75roos/i2IIxroLMDysoUEdqfZWKCXzWZrfKfR8YCviSSzIMcHS+YYUozrxHABb0hA5MfZ a5hGKQiWKCxGr7+B2rvWj9q0JQ3R74z18R+V37Z7d6dexfkKjLtZkCS8F73g4KTDOOA0BUBDhYcF D4oe3Ue09cGqEAhyOElqUFHKiKMhfvANXhAQi4Sf8g4ROBt5TyNmwSICi7BDsgtb1BDeScyEtntS vEEPI58mD8Umj94i4NM2gn5OcXOxp364As0gZn3SiNkgnANLY4d49D+8f//uZmCsrAg5BroD9HC1 xN5nkyA9r+IVxwXhBlQixAfSpqM/ug9/i5b3cbAIh8S/OyDh/3h10x+02PdhNJfLhFQXRomiI4kI 6DPmf391c93WtarFXRgvOosgbbWFvxNj31eLTFgUDYkk+JIu0JBQqP7iXSWKjmAwnTg4WVoxfQEb Q0r7K8zyT8FwBm+Ff331U6ujMwrfolkSFfQgWMURbQ7sEkivYcmTnfHfv+ixm5ve8VHv25Y9fcDZ OFotmXVBOUZO8UJdz5M7LDvne+Eg85y1xfY40Fxg33n8GWKnvvorzvoWGsiptwQ9A37UOM4z+mP2 BDQhTudUqER9GbLE+yhOAWApgYFmhwKb97cfzCZnGoWdus3okyDUQ51mfz6MpXfwHBv8jtVJWvSJ kxKPdqpDD62VVJETHOX32TkQJkZVhaR2xLjacL2GvFZsQiAxa8HokL6683Y4iiMfW24pcjerxUEa zoUSp0eH+bhJETgCiD/3vSWtjB46GTnBjWePb/olBbLuYSuJcKkHMSqtkzia00BR6RJiZQgT/ov6 uTarp6d0iRRHaCeZcVZFwTzQ1VLuEK4eGXVYb5NxyI+Tki4dzNkhU35qd8w43fHMvuXOlevR/NNw FN2Fw4XknTOjkpxEVlNRdHLEuQySn1JausfdSFpGZYcsl3uWzIxKOmagg0YnHxVYPijz4FQbVD4+ sWaHLClzZ2zSMb2eZ/Utt+CVEjUCDfQpNDIq+2YEE+ikchnVw+SyM1fof+oNymD7XEU2GOuPcfDL KliM1qKn7oz6g+aIbrwb5qqsZdQnJyZPsdGVUUnNPpxXE+s6+a9NyVgqHRlqz8FDZKH8a5poZlIh /rJuUakxqr0RQ0Voa5Q2SVuGJGc722u9FcPSgS1OTCU2t4JdLhS+FfUc+nOoWnuNGFyLpLrJG36W hKSEwQUKcHYCyJB8GGdmgVqleVPGgbf7aBtvfLZEPDfU+CyPr+a3gxqn4o90+Ic9/REKJhZXpF9H emEyjxJYsPStC7J3h7MVhZpDUeo1khPnt+ztCvqFhipaaNrAJ2EMGVSc7qzqZ7ukXD0w3zoWRK6C YZzSXUap7dI0IHgJKGdkWYt86tmCDkrJyvw4YhRukl28hdKrhApyJy3X7VAh424YqTFvrGvxmJZf w5ii//+9Rf3Pj5ItdNiK6iM/cBXY5s9YlKCXCMEEob/4jxABJXg4nTQj2asi7yurIIXxMB1qU4BV ZJ6qVTobvKlXQgAKLz0o/VHzLPBmBY90zUBSE898Y2d5lpHQByxeu1ya34o3kyQH1QMCzceggRUs RUkwW5N/ge5HEDQE+Uc3nzGcId6fZ3zykZ/Rki/nF2Z2RfOP0NdHOowHW8Bwenc85btHw/Pt5fWH yze3V9f9weX14OpycPXu+mmwOyWmPv2LXdna0/+/Ps/bNG3LtE3aqMB/+Obrk2Pr/Pe4903z/Y// Svryzn/TDa+C6ifAvdPet81V0OYE2FrWQaR2PQNGEk990VOV5bc7u/A6jaNT5d8fjhAkl31Igtv3 YFiQOrtAjRTeRlBThz8H9LfUH5GArDoJp6hHonA8oC4Jj8YRnlQt2F6CgZ57KGG0ky+0U2Gg8HMQ L4KZ0jn/mD/G3tjzvPldkGIv7Onp5uRAIaTTd6Vb2UrXBT9xIT1HnQuZh1K9LhhDeM3gn0eIirEx D0i9elJOeGi4ahZIkdJLQzfNXT0mdZvBKCt6x/F89PIKcO1kiYWmkO/K5m26ZU/NhjYBjw6BBeAh YmBszimGHi+Kkc3JIxXIOsKLlpwkt11gKYhhQc7iGeI5r7n0P7XX/Bhb/DzQn/1CP3HrsXycFu/g 3QAzPMD4YbCz0uRcKw5c/1sQR2TPuF1FoqtjWHLpNcusqe8DvAhIlkdFVVDpzapgZYhgandNafvL RsmtYJqIB99Vtq+oiPadVLKuPNwj/pavD/irC2MQPF5AXXZ/pc7cpwGoFuocWTqQQv6dSmEnqvD4 OPgFOqMNEYPTlVfPb+1rBqRWX8zaWNTXullS33PJw/6FQew5CJOvP+jqXZWR1vv7Wotn4pE2CP6I QpnBfPP1GfzqwpgKzkWCVCL4B+7O2sMDdNkLiptQHgM1W+IYfyzg/FPNu4URBZ094R7oktMtizrI /Gmo5/DvqukxFwVOV6psTZtThl4j5haU1SbhrIKsNZtZd5y0xMqCCtyIfBfZwO0RogcTuJGy2TCe kued4OWGtjtwFK0WKX6dbjREz+NDgHKr7a96TEnm7Lgc/2uFQT8DB2vy7NqnmZ7fGg72vx/+47n9 qIePNKF3tWZxLM/FLVvLu16AuvWkm6DtNLodDZPUWD79wi5yKT/Albq4ULeQn3zShVfV6s12Oz6p L0+sx23qKMwpMhkBrCLK0tCjBC9Q8Cf8DgafsFNxH0UJqnnNAjhT4MjSXUtbmw+/i/2/LW7gJnBg 5fhfR18fa/hfx4eH6P/BYo3/57+Qvjj/z65wYCffnr74Q+MD+iJ9QE8HB7a784bf7nY+5ZdTpTlB zh2j0OUER2Fge+2CODYI54h6fhckL6PZjOIFCAv+fq8UlDfL1L5AjBhjngoG3QVnjKi8W5AYJ/fh RESSnrAlxt8k2RFbBP/B+/92uyPzgoOpZemHda7bs4nHw+tjyI3DaCX6wMtS8LR/0Dts9w5bouWu J+HMtlqx8MJwBm5GuOtet3uwQ/IyRLRJtIpNSLRdaXtuQDMDf4GenqvzfBxb+0iPRiyJF6pHoDhw qV59x1Xv3QjZHQJCAVg97HMeai6H0GOVeKTauVIqfCjBa+MX7DNhfCHCF+Fv5amCDvtpwKMu4RfV OGwXleYvIBXnwBBnZAfgr45Ea+Itt8wcasYXbbTcXVcgGXPBcBcXFECGKkV37N+gPYlRW7B6iA9H RQtfolc4Klj3/EtK3qDxmqSVRaW0C5O4CBYJalAWDM4uoo+N8jYYWla+YmRZwaqBiZLWuMqmzXxf qqbPLL3BWB0VKwbtqFE1ertK/ek1XvaKaTbKbsGC2tOer1CXAW4xIN8xQvrz+CfUEvFKzA9sKnZw bY3UFklTc8mhQpJWeBPIT2oWBP8V07OAIYncABmabk6rYPEXREXmDn117wkmeVffPUGTJO6cO6Nw txkFMgiW/+KYVj4v0BJL8WuxSLcgt3vUOdSK90qL90RxTwKlCi9YUrMNvQ6MVasF1pWjzCIKk8As iMVQ7gK6CrNasjV+9BUXmjtU58BwWs3wQxEF/GHLM5nFY161vLGU6b5wzYKOjx/apRCk4lpaIXIM an9z7p/kHhOXOR+dsgNmoK82lTbnKL2ReVFgKRUl+TFKlskmm/IGTOl2VXc0OE2NulbLuOGH+sEL nVn4X8HDq+QypbMM/ZBiyXkE2a+lgoAicyYye/nMnsjEVXV5kAk8Ptm/yM2HJVKo3IxVA0bWmrIk eWgU5MDXwJVQo2uB6YcC+Ry/YeKzA/jz0/NPbJ+tn69Zq+snz5MWo//JA4g41M4dUhdr0nLWpGWs SWuyZjsepE/Eg1TnQf6LK0fHxy+Ojr/h0jN1is+0Qn6mpQI0rStB8PI/tRxNCwWpJhfFqYHBOH6z 14fFsK0WyVZL8XtqCN3UKXXTCrGblsrd9PcVvGmh5P2OTNOkdCfD13OCL9JHkHahapmblnuHI5uK bck2JB3wjJKWjc9YDt7pVQM68msNxaiOBfVzGHAlJLS2OLwhfTw+Xfvm1pWF+VtlnsBnouZXh3Db le5m0JS2XaVuWXDtMX9BxXsaUMpiGi6MuzJaTzcPGTLeE8/BlgjDFlywPVUWcK+3ORguV1RH/AtW CrhBCWIx+qc50U5K6vXyChBYjZfd3Rl1IUabmSKY28wbQvJVAHKreYRUk4lQ1PNe4oQKONuT98B/ kGV9dtI55G0bVXQQ+MI6Zkey4AW+5yTi2hQvpdGWt476yyGGyfV5aR9R+AVVJ8rx1y/OvM0hjgsA jgvgjWuCG28CcSpfqVrg0iVoqLjel+Kh1kNDhRmqA4iah0PdHAxVkw0HiG4kjx8qXhHpnMn8cQ6Q UvZ/uqLmVpE8Dqp757BXAwfYa97TVlTJXkosL2V1W9ocWE6xgrmogLh1aDxKZEIX5iu7CyZ49smX kwLs1yfUbIYTjv/swF7efZM1V7fce8ijDytfemRyMQK0i27urdSuXordKhSaomd/PIXi0ZTjhrlm akYqW3c+hMn5hLRTvij55oAxhJqWJfyRx5iGp/v7TLfJ8iRG0YyTwB95EvBUkZBEDMRqGoxwQ/HM cOLzjl0QTVk3q+2u3ztT2Y/i1/4+5cqMx8wUdTItmkxgNrq/kmetmGnloNwGu0rGafRtN0vPM0CW 7bNjfgaMZSZ8VLJHL+iLNfxc+HGn9r1K5GZbWygEZi55Wfh3Atyn/2TFi7+591b80QGNKU79vTdR tBQxryYQMjEEePExtRj3MR3boY1jWTiPaSwlRZg+tGgAe4G7Z+KP8wv2Qv7eR5D2kyJRoYmhglw6 bMzry8VYg6vWarYlu3XgauYEgBZ9AgZ8TG3Uavkoo4FdOj/XBiydVIEEpFahDeouTeLsu95d6Zdz w3Ab43IM59Ga5miZzbKecxMgsNdum5KnD3UeJiPoMynY2WC3p21L5DXeNcGwY4r7pBhOm01Y5N3k JRaAPd6URXQdWgSjjm4J0KV6ATnucUejTtptEFS3wtUeMrgwB00g0Uq+kTIjqLoloSrVa6rUeLLb 8mwUda1lLkBiGfoz3RuTLhDN+3Qf8CwTqt4sjrKIlDS1zCqg0RFl1YOSnlkI7/EaIdgJynII0ith 2e/WbD78GZ9K1qVRxK9/7Tm4bc1EXUv2sHN42FOdj9diYj5bE2RqjTzPHMZfTGj98tXEQcDqP64b +C+h2zvB7dmzII5bVn/NPr0kZHt+I0lWzqHoOyoKwHtnl+gn/B+PIlKHI2Ej7mc+gU1lYqK9xdcf 3rzZO7OJGNtegSMFazLhIl1rm5x72nec9LIpVxsFTTcrme+sk1vMdY2ZVqqn8+sFzbfHm9SkJjWp SU1qUpOa1KQmNalJTWpSk5rUpCY1qUlNalKTmtSkJjWpSU1qUpOa1KQmNalJTWrSl5r+A965eYUA GAEA --++----------20040113203926-074758809----------++-- From luis.ibanez@kitware.com Thu Jan 15 22:10:31 2004 From: luis.ibanez@kitware.com (Luis Ibanez) Date: Thu, 15 Jan 2004 17:10:31 -0500 Subject: [Insight-developers] correction In-Reply-To: References: Message-ID: <40070FD7.8030209@kitware.com> Hi Samson, Thanks for posting the corrections. This issue was entered as Bug # 519. The files and classes were renamed according to your indications. The content of the algorithm was not updated yet from your files. Since this is a separate issue. Once we verify that the renaming passed fine in the dashboard we will update the content of the algorithm from the files in your web site. Regards, Luis BTW your attachement was expanded as text in your image, so we will use your web link. ---------------------- Samson Timoner wrote: > Content-Type: multipart/mixed; > boundary="++----------20040113203926-074758809----------++" > > This is a multi-part message in MIME format. > --++----------20040113203926-074758809----------++ > Content-Type: text/plain; charset=us-ascii > Content-Transfer-Encoding: 7bit > > > When I submitted itkKLCompareHistogramImageToImageMetric, its name was > changed to itkKullbackLieblerCompareHistogramImageToImageMetric. Alas, > "Liebler" is a mis-spelling of "Leibler". I noticed this last week, > and we should probably fix it before the next release. All the changes > are in files that are attached. I submitted an experimental build > from horus0.ai.mit.edu with the new files. > > The following files should be removed from CVS: > Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetric.h > Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetric.txx > Testing/Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetricTes > t.cxx > > The following files should be added to CVS: > Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetric.h > Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetric.txx > Testing/Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetricTes > t.cxx > > The following files should be changed, and are attached: > Testing/Code/Algorithms/CMakeLists.txt > Testing/Code/Algorithms/itkAlgorithmsHeaderTest.cxx > Testing/Code/Algorithms/itkAlgorithmsTests2.cxx > Testing/Code/Algorithms/itkCompareHistogramImageToImageMetricTest.cxx > > Note that I also made the following changes: > 1) I added some comments to itkKullbackLeiblerCompareHistogramImageToImageMetr > ic. > 2) While my implementation of the algorithm was OK, it in fact did not > correspond exactly to the paper I referenced. I therefore changed > the code to make it do so. This changed 2 lines of code in > itkKullbackLeiblerCompareHistogramImageToImageMetric.txx > 3) I changed itkKullbackLeiblerCompareHistogramImageToImageMetricTest.cxx > so that it would have a minimum at the appropriate location using > the changed algorithm. > > I'm sorry to have caused problems, > > -- Samson > > P.S. If you do not like attachments, I put all the files in: > http://www.ai.mit.edu/people/samson/upload/itk/ > > --++----------20040113203926-074758809----------++ > Content-Type: application/x-gunzip; name="submit.tar.gz" > Content-Transfer-Encoding: base64 > > H4sICGiVBEAAA3N1Ym1pdC50YXIA7F19c9s20s/f+hS4tHMjp471YjtOnbQzsl5sTSTbI8lxO70b From hans-johnson@uiowa.edu Fri Jan 16 13:47:12 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Fri, 16 Jan 2004 07:47:12 -0600 Subject: [Insight-developers] Failed Tests Report From Iowa Message-ID: <1074260832.31496.46.camel@ipl-imac.psychiatry.uiowa.edu> Hello All, I know that a new release is going to be made very soon, so I am going to try to summarize the test failures at the Iowa site. Unfortunatly, I will not have much time to be able to track these down myself or work on this much until after Feb. 1. ================================================ IRIX64-CC-7.3-dbg Testing-- ******vnl_test_math Bug report already submitted, and e-mail sent. Brad King is looking into it (see developer list from Monday Jan-12-2004) ******itkKullbackLeiblerCompareHistogramImageToImageMetricTest No test is not registered. ******Tcl wrapping tests Wrapping is failing. Building-- Wrapping is failing. My guess is that this is that some test is missing in detecting the build environment, but I have not been able to track it down. CMakeCache.txt/CMakeError.txt/itkConfigure.h/ ================================================ Linux-icc-7.1-opt Testin--- ******itkKullbackLeiblerCompareHistogramImageToImageMetricTest and itkMutialInformationMetricTest, Exception handling not done correctly for this NULL Pointer (both fail for "Try causing a exception by making fixed image NULL" test) ================================================ Linux-g++-3.2-opt *****ImageRegistration4Test--- Huge errors in difference images -- This has been occuring for quite a long time--Before Jan, 1, 2004. *****MultiResImageRegistration2Test--- Errors in difference images. -- This has been occuring for quite a long time--Before Jan, 1, 2004. Regards, Hans From luis.ibanez@kitware.com Fri Jan 16 16:57:41 2004 From: luis.ibanez@kitware.com (Luis Ibanez) Date: Fri, 16 Jan 2004 11:57:41 -0500 Subject: [Insight-developers] Failed Tests Report From Iowa In-Reply-To: <1074260832.31496.46.camel@ipl-imac.psychiatry.uiowa.edu> References: <1074260832.31496.46.camel@ipl-imac.psychiatry.uiowa.edu> Message-ID: <40081805.7040502@kitware.com> Hi Hans, The test: itkKullbackLeiblerCompareHistogramImageToImageMetricTest is registered in Testing/Code/Algorithms/itkAlgorithmsTests2.cxx It may seem that the source code in IRIX64-CC-7.3-dbg is out of sync with CVS.... The filter was renamed yesterday.... Luis ---------------------- Hans J. Johnson wrote: > Hello All, > > I know that a new release is going to be made very soon, so I am going > to try to summarize the test failures at the Iowa site. Unfortunatly, I > will not have much time to be able to track these down myself or work on > this much until after Feb. 1. > > ================================================ > IRIX64-CC-7.3-dbg > Testing-- > ******vnl_test_math > Bug report already submitted, and e-mail sent. Brad King is looking > into it (see developer list from Monday Jan-12-2004) > ******itkKullbackLeiblerCompareHistogramImageToImageMetricTest > No test is not registered. > ******Tcl wrapping tests > Wrapping is failing. > Building-- > Wrapping is failing. My guess is that this is that some test is > missing in detecting the build environment, but I have not been able to > track it down. CMakeCache.txt/CMakeError.txt/itkConfigure.h/ > > > ================================================ > Linux-icc-7.1-opt > Testin--- > ******itkKullbackLeiblerCompareHistogramImageToImageMetricTest and > itkMutialInformationMetricTest, > Exception handling not done correctly for this NULL Pointer (both > fail for "Try causing a exception by making fixed image NULL" test) > > ================================================ > Linux-g++-3.2-opt > *****ImageRegistration4Test--- > Huge errors in difference images -- This has been occuring for quite > a long time--Before Jan, 1, 2004. > *****MultiResImageRegistration2Test--- > Errors in difference images. -- This has been occuring for quite a > long time--Before Jan, 1, 2004. > > Regards, > Hans > > > > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers > From luis.ibanez@kitware.com Mon Jan 19 17:40:04 2004 From: luis.ibanez@kitware.com (Luis Ibanez) Date: Mon, 19 Jan 2004 12:40:04 -0500 Subject: [Insight-developers] correction In-Reply-To: References: Message-ID: <400C1674.6070400@kitware.com> Hi Samson, The implementation of the Kullback Leibler classes has been updated from the web site you posted. This closes Bug #520. Please let us know if you find further problems, Thanks Luis ----------------------- Samson Timoner wrote: > > When I submitted itkKLCompareHistogramImageToImageMetric, its name was > changed to itkKullbackLieblerCompareHistogramImageToImageMetric. Alas, > "Liebler" is a mis-spelling of "Leibler". I noticed this last week, > and we should probably fix it before the next release. All the changes > are in files that are attached. I submitted an experimental build > from horus0.ai.mit.edu with the new files. > > The following files should be removed from CVS: > Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetric.h > Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetric.txx > Testing/Code/Algorithms/itkKullbackLieblerCompareHistogramImageToImageMetricTes > t.cxx > > The following files should be added to CVS: > Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetric.h > Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetric.txx > Testing/Code/Algorithms/itkKullbackLeiblerCompareHistogramImageToImageMetricTes > t.cxx > > The following files should be changed, and are attached: > Testing/Code/Algorithms/CMakeLists.txt > Testing/Code/Algorithms/itkAlgorithmsHeaderTest.cxx > Testing/Code/Algorithms/itkAlgorithmsTests2.cxx > Testing/Code/Algorithms/itkCompareHistogramImageToImageMetricTest.cxx > > Note that I also made the following changes: > 1) I added some comments to itkKullbackLeiblerCompareHistogramImageToImageMetr > ic. > 2) While my implementation of the algorithm was OK, it in fact did not > correspond exactly to the paper I referenced. I therefore changed > the code to make it do so. This changed 2 lines of code in > itkKullbackLeiblerCompareHistogramImageToImageMetric.txx > 3) I changed itkKullbackLeiblerCompareHistogramImageToImageMetricTest.cxx > so that it would have a minimum at the appropriate location using > the changed algorithm. > > I'm sorry to have caused problems, > > -- Samson > > P.S. If you do not like attachments, I put all the files in: > http://www.ai.mit.edu/people/samson/upload/itk/ > From brad.king@kitware.com Mon Jan 19 18:04:45 2004 From: brad.king@kitware.com (Brad King) Date: Mon, 19 Jan 2004 13:04:45 -0500 (EST) Subject: [Insight-developers] james.uiowa Linux-icc-7.1-opt Message-ID: Hans, There are two tests failing on your Linux-icc-7.1-opt build. I've built a copy of ITK using the same compiler version and C/CXX flags, but the tests do not fail for me. The two tests fail the exact same way. They print this: "Try causing a exception by making fixed image NULL" and then crash. On my build the output continues with "Caught the exception." Any idea what could be different on that machine? Thanks, -Brad From hans-johnson@uiowa.edu Mon Jan 19 18:44:52 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Mon, 19 Jan 2004 12:44:52 -0600 Subject: [Insight-developers] Optimization problems with gcc3 Message-ID: <1074537891.26799.34.camel@ipl-imac.psychiatry.uiowa.edu> Hello All, I just wanted to let everyone know of some testing I've done in trying to get the tesla.uiowa Linux-gcc-3.2-opt to test without errors. It looks like the -O3 optimizations for the gcc compiler are causing the test to fail. I'm not 100% sure this is a compiler bug, but I've changed my nightly build compile flags anyway. FAILS1---"-DNDEBUG -O3 -funroll-loops -fno-omit-frame-pointer -fstrict-aliasing -ftemplate-depth-50" FAILS2---"-DNDEBUG -O2 -frename-registers -finline-functions -funroll-loops -fno-omit-frame-pointer -fstrict-aliasing -ftemplate-depth-50" FAILS3---"-DNDEBUG -O2 -finline-functions -funroll-loops -fno-omit-frame-pointer -fstrict-aliasing -ftemplate-depth-50" SUCCESS--"-DNDEBUG -O2 -frename-registers -funroll-loops -fno-omit-frame-pointer -fstrict-aliasing -ftemplate-depth-50" NOTE: "-O3" == "-O2 -frename-registers -finline-functions" =====Output from Success=========== ============Output from FAILED3======== 0 -0.321485 [3.97061, 0.484012] 0 -0.321485 [3.97061, 0.484012] 1 -0.360223 [7.45659, -1.4776] 1 -0.360223 [7.45659, -1.4776] 2 -0.361493 [6.82751, 0.42089] 2 -0.312199 [6.86882, 0.434084] 3 -0.373211 [7.82578, 0.479614] 3 -0.365193 [7.86822, 0.468775] 4 -0.382745 [8.77673, 0.170261] 4 -0.345765 [8.80597, 0.121454] 5 -0.383594 [9.76414, 0.328419] 5 -0.358856 [9.74668, 0.460672] 6 -0.391406 [10.7634, 0.28909] 6 -0.344188 [10.7464, 0.483857] 7 -0.396325 [11.7548, 0.158644] 7 -0.383695 [11.7294, 0.299962] 8 -0.395927 [12.3374, 0.971434] 8 -0.38631 [12.0244, 1.25544] 9 -0.400565 [11.8154, 1.82437] 9 -0.3651 [11.7557, 0.833794] 10 -0.405265 [11.9641, 2.81324] 10 -0.39056 [11.8499, 1.06538] 11 -0.412902 [12.5275, 3.63948] 11 -0.365319 [12.0709, 1.18225] 12 -0.425711 [13.0856, 4.46919] 12 -0.377377 [11.9752, 1.10175] 13 -0.402132 [13.9658, 4.94392] 13 -0.378953 [12.0329, 1.12581] 14 -0.430453 [13.6788, 5.35335] 14 -0.330904 [12.09, 1.15136] 15 -0.441748 [13.212, 5.17408] 15 -0.314393 [12.0804, 1.1216] 16 -0.437343 [13.1785, 5.42182] 16 -0.380332 [12.05, 1.11429] 17 -0.443244 [13.1794, 5.67182] 17 -0.391133 [12.0602, 1.12616] 18 -0.447031 [13.1572, 5.92083] 18 -0.2877 [12.0756, 1.12899] 19 -0.445267 [13.0639, 6.15278] 19 -0.347698 [12.0868, 1.11811] 20 -0.44797 [13.025, 6.39973] 20 -0.387801 [12.0795, 1.11522] 21 -0.45237 [13.1422, 6.35621] Result = 22 -0.453201 [13.1234, 6.41582] Translation X = 12.0795 23 -0.453964 [13.1084, 6.4765] Translation Y = 1.11522 24 -0.442526 [13.1389, 6.46956] Iterations = 22 25 -0.454963 [13.1482, 6.49939] Metric value = -0.336555 26 -0.455464 [13.147, 6.48381] Stop Condition = 2 27 -0.442931 [13.1517, 6.49007] Result = Translation X = 13.1517 Translation Y = 6.49007 Iterations = 29 Metric value = -0.454962 Stop Condition = 2 ================================== =========================================== Regards, Hans From hans-johnson@uiowa.edu Tue Jan 20 02:40:58 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Mon, 19 Jan 2004 20:40:58 -0600 Subject: [Insight-developers] james.uiowa Linux-icc-7.1-opt In-Reply-To: References: Message-ID: <1074566458.21915.17.camel@ipl-imac.psychiatry.uiowa.edu> Brad, I've found a way to clean the last two testing errors for the intel compilers. This may be a compiler optimization error, but it is easy to work around. The segfault occurs in the itkExceptionMacro. The segfault occurs during the destruction of an std::basic_string (I'm guessing it's the message string, but I'm not sure). I figured out how to side-step the problem while trying to debug the program. My guess is that the problem occurs because their is an object creation, an object destruction all within the throw scopping. If the itkExceptionMacro is made more explicit, the error goes away. Here is the new itkExceptionMacro: { ::itk::OStringStream message; message << "itk::ERROR: " << this->GetNameOfClass() << "(" << this << "): " x ; ::itk::ExceptionObject tempException(__FILE__, __LINE__, message.str().c_str());//Explicit naming to avoid compiler error on intel throw tempException; } I did not make this change, so if you could to it, I would really appreciate it. Thanks, Hans PS: This may also clear the G5 errors. On Mon, 2004-01-19 at 12:04, Brad King wrote: > Hans, > > There are two tests failing on your Linux-icc-7.1-opt build. I've built a > copy of ITK using the same compiler version and C/CXX flags, but the tests > do not fail for me. The two tests fail the exact same way. They print > this: > > "Try causing a exception by making fixed image NULL" > > and then crash. On my build the output continues with > > "Caught the exception." > > Any idea what could be different on that machine? > > Thanks, > -Brad > > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers From brad.king@kitware.com Tue Jan 20 13:58:07 2004 From: brad.king@kitware.com (Brad King) Date: Tue, 20 Jan 2004 08:58:07 -0500 (EST) Subject: [Insight-developers] james.uiowa Linux-icc-7.1-opt In-Reply-To: <1074566458.21915.17.camel@ipl-imac.psychiatry.uiowa.edu> Message-ID: On Mon, 19 Jan 2004, Hans J. Johnson wrote: > If the itkExceptionMacro is made more explicit, the error goes away. > > Here is the new itkExceptionMacro: > { > ::itk::OStringStream message; > message << "itk::ERROR: " << this->GetNameOfClass() > << "(" << this << "): " x ; > ::itk::ExceptionObject tempException(__FILE__, __LINE__, > message.str().c_str());//Explicit naming to avoid compiler error on > intel > throw tempException; > } > > I did not make this change, so if you could to it, I would really > appreciate it. I've made a change using this technique. Thanks alot for tracking this down. -Brad From hjohnson@mail.psychiatry.uiowa.edu Wed Jan 21 01:26:04 2004 From: hjohnson@mail.psychiatry.uiowa.edu (Hans Johnson) Date: Tue, 20 Jan 2004 19:26:04 -0600 Subject: [Insight-developers] ITK Dartboard not working Message-ID: <400DD52C.9070607@mail.psychiatry.uiowa.edu> Hello Darboard maintainers, It looks like the dart board is not working for ITK. The last time this happend, it was because web server ran out of disk space. Hans From lorensen@crd.ge.com Wed Jan 21 01:31:17 2004 From: lorensen@crd.ge.com (Lorensen, William E (Research)) Date: Tue, 20 Jan 2004 20:31:17 -0500 Subject: [Insight-developers] ITK Dartboard not working Message-ID: Seems to be fine here. What symptoms are you seeing? -----Original Message----- From: Hans Johnson [mailto:hjohnson@mail.psychiatry.uiowa.edu] Sent: Tuesday, January 20, 2004 8:26 PM To: insight-developers@itk.org Subject: [Insight-developers] ITK Dartboard not working Hello Darboard maintainers, It looks like the dart board is not working for ITK. The last time this happend, it was because web server ran out of disk space. Hans _______________________________________________ Insight-developers mailing list Insight-developers@itk.org http://www.itk.org/mailman/listinfo/insight-developers From hans-johnson@uiowa.edu Wed Jan 21 15:19:45 2004 From: hans-johnson@uiowa.edu (Hans J. Johnson) Date: Wed, 21 Jan 2004 09:19:45 -0600 Subject: [Insight-developers] SGI WRAPPING WORKS Message-ID: <1074698385.7082.1.camel@ipl-imac.psychiatry.uiowa.edu> Hello All, I just noticed that the dgb version of the SGI build ran and passed all it's tests-- INCLUDING WRAPPING!! The opt version is still chugging away, but it seem OK so far. I would like to give a big thank-you to Brad King, you help on this has been great. Thanks, Hans From mole_sidahmed@web.de Thu Jan 22 15:14:02 2004 From: mole_sidahmed@web.de (Ould Sid'Ahmed Moh. Lemine) Date: Thu, 22 Jan 2004 16:14:02 +0100 Subject: [Insight-developers] Detection of composite edges Message-ID: <200401221514.i0MFE2Q06842@mailgate5.cinetic.de>

Hi! insight-devolopers,

I had a requirement to create a class - methodes -,which detects composite edges in an image.Has ITK any class or functions, that surves my requirement?Please could you suggest any solution.

Thanks. Ould Sid-Ahmed

______________________________________________________________________________
Erdbeben im Iran: Zehntausende Kinder brauchen Hilfe. UNICEF hilft den
Kindern - helfen Sie mit! https://www.unicef.de/spe/spe_03.php From luis.ibanez@kitware.com Thu Jan 22 15:46:49 2004 From: luis.ibanez@kitware.com (Luis Ibanez) Date: Thu, 22 Jan 2004 10:46:49 -0500 Subject: [Insight-developers] Detection of composite edges In-Reply-To: <200401221514.i0MFE2Q06842@mailgate5.cinetic.de> References: <200401221514.i0MFE2Q06842@mailgate5.cinetic.de> Message-ID: <400FF069.7040204@kitware.com> Hi Ould Sid-Ahmed, Could you please share with us your definition of "composite edge" ? Also, you may want to use the users-list instead of the developers-list, unless your main purpose is to contribute your code to ITK. Regards, Luis ----------------------------------- Ould Sid'Ahmed Moh. Lemine wrote: > Hi! insight-devolopers, > > I had a requirement to create a class - methodes -,which detects composite edges in an image.Has ITK any class or functions, that surves my requirement?Please could you suggest any solution. > > Thanks. Ould Sid-Ahmed > > ______________________________________________________________________________ > Erdbeben im Iran: Zehntausende Kinder brauchen Hilfe. UNICEF hilft den > Kindern - helfen Sie mit! https://www.unicef.de/spe/spe_03.php > _______________________________________________ Insight-developers > mailing list Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers From brad.king@kitware.com Thu Jan 22 15:59:00 2004 From: brad.king@kitware.com (Brad King) Date: Thu, 22 Jan 2004 10:59:00 -0500 (EST) Subject: [Insight-developers] ITK 1.6 Branched Message-ID: Hello, I have branched Insight for the 1.6 release (I actually branched yesterday but have been working on a few problems). Developers are now free to commit new changes to the main tree in CVS. The release branch tag is "ITK-1-6". I have not yet created a 1.6.0 sticky tag on the branch. I'm currently configuring a few machines to perform dashboards for the 1.6 branch. -Brad From norman-k-williams@uiowa.edu Thu Jan 22 19:26:10 2004 From: norman-k-williams@uiowa.edu (Kent Williams) Date: 22 Jan 2004 13:26:10 -0600 Subject: [Insight-developers] SNAP application link failure Message-ID: <1074799570.11284.3.camel@nissl.psychiatry.uiowa.edu> There seems to be a collision in linking the SNAP sample application, whereby it tries to pull in the PNG code from both VTK's libvtkpng.a and ITK's libitkpng.a. Apparently both versions of png.o get loaded at link, causing multiply defined symbol errors with GCC3 on Linux. I don't use this software, but I do try and compile all the applications once a week or so ;-) From brad.king@kitware.com Thu Jan 22 22:09:00 2004 From: brad.king@kitware.com (Brad King) Date: Thu, 22 Jan 2004 17:09:00 -0500 (EST) Subject: [Insight-developers] SNAP application link failure In-Reply-To: <1074799570.11284.3.camel@nissl.psychiatry.uiowa.edu> Message-ID: On 22 Jan 2004, Kent Williams wrote: > There seems to be a collision in linking the SNAP sample application, > whereby it tries to pull in the PNG code from both VTK's libvtkpng.a and > ITK's libitkpng.a. > > Apparently both versions of png.o get loaded at link, causing multiply > defined symbol errors with GCC3 on Linux. > > I don't use this software, but I do try and compile all the applications > once a week or so ;-) This was fixed a couple weeks ago by mangling the names in ITK's version of the library. Run "make clean" in the Insight-build/Utilities directory and then rebuild everything. -Brad From shivak@vitinfotech.com Fri Jan 23 12:31:19 2004 From: shivak@vitinfotech.com (Shiva Kumar) Date: Fri, 23 Jan 2004 18:01:19 +0530 Subject: [Insight-developers] Regarding using RGB colors with pixels in ITKImage Message-ID: <001801c3e1ac$cff23f50$867ba8c0@vit.com> This is a multi-part message in MIME format. ------=_NextPart_000_0015_01C3E1DA.E6F19CE0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable hi everyOne, For creating an Image using Itk I am defining image as=20 for ex: typedef unsigned int PixelType; typedef itk::Image < PixelType, 2 > ImageType; here we are using unsigned int but how to associate a color with pixel?. there is another template class for specifying pixelType like: typedef itk::RGBPixel< Unsigned char> pixelType typedef itk::Image < PixelType, 2 > ImageType This works fine but the problem is I am using a filter = itktovtkImageFilter to convert into vtkImage for rendering but this filter will not accept RGBPixeltype. Then if i use unsigned int or unsigned char how can i specify the color = values. How does the filter knows about color representation? Please anybody has the answer let me know ASAP. Thanks and Regds Shiva kumar=20 ------=_NextPart_000_0015_01C3E1DA.E6F19CE0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

hi everyOne,

For creating an = Image using=20 Itk I am defining image as

for ex:

typedef =20 unsigned int PixelType;
typedef = itk::Image <=20 PixelType, 2 > ImageType;

here we are using unsigned int but how = to associate=20 a color with pixel?.

there is another template class for = specifying=20 pixelType like:

= typedef itk::RGBPixel< =20 Unsigned char> pixelType
typedef itk::Image < = PixelType, 2=20 > ImageType

This works fine but the problem is I am = using a=20 filter itktovtkImageFilter to convert into vtkImage for=20 rendering

but this=20 filter will not accept RGBPixeltype.

Then if i use unsigned int or unsigned = char how can=20 i specify the color values.

How does the filter knows about color=20 representation?

Please anybody has the answer let = me know=20 ASAP.

Thanks and Regds

Shiva kumar

------=_NextPart_000_0015_01C3E1DA.E6F19CE0-- From hjohnson@mail.psychiatry.uiowa.edu Sun Jan 25 21:53:52 2004 From: hjohnson@mail.psychiatry.uiowa.edu (Hans Johnson) Date: Sun, 25 Jan 2004 15:53:52 -0600 Subject: [Insight-developers] GCC 2.96 wrapping failures Message-ID: <40143AF0.5050006@mail.psychiatry.uiowa.edu> Brad or other CableSwig Experts, CableSwig does not work under a redhat 7.3 gcc 2.96 installation of the compiler. I've pasted the errors below, any suggested would be greatly appreciated. Thanks, Hans Building wrap_ITKCommonBase.xml from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight-build/Wrapping/CSwig/Common/wrap_ITKCommonBase.xml... In file included from /usr/include/g++-3/stl_alloc.h:71, from /usr/include/g++-3/alloc.h:21, from /usr/include/g++-3/std/bastring.h:39, from /usr/include/g++-3/string:6, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Utilities/CableSwig/GCC_XML/Support/GCC/2.95/string:4, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkMacro.h:36, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkSmartPointer.h:20, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkLightObject.h:27, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkObject.h:23, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkCommand.h:20, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Wrapping/CSwig/Common/wrap_ITKCommonBase.cxx:21: /usr/include/pthread.h:163: error: syntax error before `__thread' /usr/include/pthread.h:165: error: `pthread_create' declared as function returning a function /usr/include/pthread.h:166: error: syntax error before `void' /usr/include/pthread.h:591: error: storage class specifiers invalid in parameter declarations /usr/include/pthread.h:591: error: storage class specified for parameter ` parameter' In file included from /usr/include/pthread.h:655, from /usr/include/g++-3/stl_alloc.h:71, from /usr/include/g++-3/alloc.h:21, from /usr/include/g++-3/std/bastring.h:39, from /usr/include/g++-3/string:6, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Utilities/CableSwig/GCC_XML/Support/GCC/2.95/string:4, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkMacro.h:36, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkSmartPointer.h:20, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkLightObject.h:27, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkObject.h:23, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkCommand.h:20, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Wrapping/CSwig/Common/wrap_ITKCommonBase.cxx:21: /usr/include/bits/sigthread.h:36: error: storage class specifiers invalid in parameter declarations /usr/include/bits/sigthread.h:36: error: storage class specified for parameter `parameter' In file included from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Utilities/CableSwig/GCC_XML/Support/GCC/2.95/sstream:4, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkMacro.h:40, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkSmartPointer.h:20, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkLightObject.h:27, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkObject.h:23, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Code/Common/itkCommand.h:20, from /home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight/Wrapping/CSwig/Common/wrap_ITKCommonBase.cxx:21: /usr/include/g++-3/sstream:260: error: default argument given for parameter 3 of `streampos std::stringbuf::seekoff(long long int, ios::seek_dir, int)' /usr/include/g++-3/sstream:83: error: after previous specification in `virtual streampos std::stringbuf::seekoff(long long int, ios::seek_dir, int)' /usr/include/g++-3/sstream:278: error: default argument given for parameter 2 of `streampos std::stringbuf::seekpos(long long int, int)' /usr/include/g++-3/sstream:84: error: after previous specification in `virtual streampos std::stringbuf::seekpos(long long int, int)' gmake[1]: *** [/home/hjohnson/src/brains2/iplFreeware/ix86/PROFILE/Insight-build/Wrapping/CSwig/Common/wrap_ITKCommonBase.xml] Error 1 gmake: *** [default_target] Error 2 From jjomier@cs.unc.edu Mon Jan 26 00:10:53 2004 From: jjomier@cs.unc.edu (Julien Jomier) Date: Sun, 25 Jan 2004 19:10:53 -0500 Subject: [Insight-developers] itkIPLCommonImageIO MSVC Warning 4284 Message-ID: <001b01c3e3a0$deeaace0$5d691398@pissarro> Hi, This is not the first time I'm getting this warning with MSVC 6 when compiling ITK: [snip] itkIPLCommonImageIO.cxx C:\Program Files\Microsoft Visual Studio\VC98\INCLUDE\list(68) : warning C4284: return type for 'std::list >::const_iterator::operator = ->' is 'class itk::IPLFileSortInfo *const * ' (ie; not a UDT or reference to a UDT. Will produce errors = if applied using infix notation) E:\Workspace\Insight\Code\IO\itkIPLFileNameList.h(98) : see reference to class template instantiation 'std::list >' being compiled C:\Program Files\Microsoft Visual Studio\VC98\INCLUDE\list(103) : = warning C4284: return type for 'std::list >::iterator::operator ->' = is 'class itk::IPLFileSortInfo ** ' (ie; not a UDT or reference to a UDT. Will produce errors if applied = using infix notation) E:\Workspace\Insight\Code\IO\itkIPLFileNameList.h(98) : see reference to class template instantiation 'std::list >' being compiled [snip] Should we disable this warning (4284) for MSVC or try to fix it instead? Julien From norman-k-williams@uiowa.edu Mon Jan 26 05:15:13 2004 From: norman-k-williams@uiowa.edu (kent williams) Date: Sun, 25 Jan 2004 23:15:13 -0600 Subject: [Insight-developers] itkIPLCommonImageIO MSVC Warning 4284 References: <001b01c3e3a0$deeaace0$5d691398@pissarro> Message-ID: <006001c3e3cb$60f9b670$6a00a8c0@LAPCRUSHER> I'm not sure how one would fix this problem, as the warning arises because of Visual Studio's confusion when it elaborates the std::list for IPLFileSortInfo *. It seems to be saying that a list must be either of objects, or of references to objects, when it's obvious that a pointer to an object is a perfectly fine thing to keep in a std::list. I could revisit this code and change it to a list of references -- it'd be easy enough. As of now we're not building with Visual Studio at Iowa, though, so I'd only find out second-hand if we've muzzled MSVC6. A few references to this error message -- mostly disparaging: http://www.squid-cache.org/mail-archive/squid-dev/200307/0067.html Here's a page that explicitly recommends disabling that warning with pragmas; apparently this warning shows up frequently if you do anything with STL and Visual C++. Several Free/GPL packages whose source code is indexed by Google do exactly this as a matter of course: http://www.codeproject.com/vcpp/stl/stl_without_warnings.asp?print=true ----- Original Message ----- From: "Julien Jomier" To: Sent: Sunday, January 25, 2004 6:10 PM Subject: [Insight-developers] itkIPLCommonImageIO MSVC Warning 4284 > Hi, > > This is not the first time I'm getting this warning with MSVC 6 when > compiling ITK: > > [snip] > > itkIPLCommonImageIO.cxx > C:\Program Files\Microsoft Visual Studio\VC98\INCLUDE\list(68) : warning > C4284: return type for 'std::list std::allocator >::const_iterator::operator ->' > is 'class itk::IPLFileSortInfo > *const * ' (ie; not a UDT or reference to a UDT. Will produce errors if > applied using infix notation) > E:\Workspace\Insight\Code\IO\itkIPLFileNameList.h(98) : see > reference to class template instantiation 'std::list itk::IPLFileSortInfo *,class std::allocator >' > being compiled > C:\Program Files\Microsoft Visual Studio\VC98\INCLUDE\list(103) : warning > C4284: return type for 'std::list std::allocator >::iterator::operator ->' is > 'class itk::IPLFileSortInfo ** ' > (ie; not a UDT or reference to a UDT. Will produce errors if applied using > infix notation) > E:\Workspace\Insight\Code\IO\itkIPLFileNameList.h(98) : see > reference to class template instantiation 'std::list itk::IPLFileSortInfo *,class std::allocator >' > being compiled > > [snip] > > Should we disable this warning (4284) for MSVC or try to fix it instead? > > Julien > > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers > From lorensen@crd.ge.com Mon Jan 26 12:59:50 2004 From: lorensen@crd.ge.com (Lorensen, William E (Research)) Date: Mon, 26 Jan 2004 07:59:50 -0500 Subject: [Insight-developers] itkIPLCommonImageIO MSVC Warning 4284 Message-ID: I added a pragma in the IPLCommonImageIO.cxx file to block the warning. -----Original Message----- From: kent williams [mailto:norman-k-williams@uiowa.edu] Sent: Monday, January 26, 2004 12:15 AM To: Julien Jomier; insight-developers@public.kitware.com Subject: Re: [Insight-developers] itkIPLCommonImageIO MSVC Warning 4284 I'm not sure how one would fix this problem, as the warning arises because of Visual Studio's confusion when it elaborates the std::list for IPLFileSortInfo *. It seems to be saying that a list must be either of objects, or of references to objects, when it's obvious that a pointer to an object is a perfectly fine thing to keep in a std::list. I could revisit this code and change it to a list of references -- it'd be easy enough. As of now we're not building with Visual Studio at Iowa, though, so I'd only find out second-hand if we've muzzled MSVC6. A few references to this error message -- mostly disparaging: http://www.squid-cache.org/mail-archive/squid-dev/200307/0067.html Here's a page that explicitly recommends disabling that warning with pragmas; apparently this warning shows up frequently if you do anything with STL and Visual C++. Several Free/GPL packages whose source code is indexed by Google do exactly this as a matter of course: http://www.codeproject.com/vcpp/stl/stl_without_warnings.asp?print=true ----- Original Message ----- From: "Julien Jomier" To: Sent: Sunday, January 25, 2004 6:10 PM Subject: [Insight-developers] itkIPLCommonImageIO MSVC Warning 4284 > Hi, > > This is not the first time I'm getting this warning with MSVC 6 when > compiling ITK: > > [snip] > > itkIPLCommonImageIO.cxx > C:\Program Files\Microsoft Visual Studio\VC98\INCLUDE\list(68) : warning > C4284: return type for 'std::list std::allocator >::const_iterator::operator ->' > is 'class itk::IPLFileSortInfo > *const * ' (ie; not a UDT or reference to a UDT. Will produce errors if > applied using infix notation) > E:\Workspace\Insight\Code\IO\itkIPLFileNameList.h(98) : see > reference to class template instantiation 'std::list itk::IPLFileSortInfo *,class std::allocator >' > being compiled > C:\Program Files\Microsoft Visual Studio\VC98\INCLUDE\list(103) : warning > C4284: return type for 'std::list std::allocator >::iterator::operator ->' is > 'class itk::IPLFileSortInfo ** ' > (ie; not a UDT or reference to a UDT. Will produce errors if applied using > infix notation) > E:\Workspace\Insight\Code\IO\itkIPLFileNameList.h(98) : see > reference to class template instantiation 'std::list itk::IPLFileSortInfo *,class std::allocator >' > being compiled > > [snip] > > Should we disable this warning (4284) for MSVC or try to fix it instead? > > Julien > > _______________________________________________ > Insight-developers mailing list > Insight-developers@itk.org > http://www.itk.org/mailman/listinfo/insight-developers > _______________________________________________ Insight-developers mailing list Insight-developers@itk.org http://www.itk.org/mailman/listinfo/insight-developers From brad.king@kitware.com Wed Jan 28 19:39:50 2004 From: brad.king@kitware.com (Brad King) Date: Wed, 28 Jan 2004 14:39:50 -0500 (EST) Subject: [Insight-developers] Re: [Insight-users] TODO for AffineTransform::Rotate3D() In-Reply-To: <4017E90B.2080006@kitware.com> Message-ID: On Wed, 28 Jan 2004, Luis Ibanez wrote: > The concept checking for dimension seems to be ok when we only use C++, > since the methods are only compiled when used, but... wouldn't it > generate problems for wrapping ? We are instantiating this transform for > both 2D and 3D and the wrapping layer instantiates all the methods. They are only compiled when used if we are not doing explicit instantiation. An explicit instantiation will compile all methods. Since I'm looking into using explicit instantiation for common types to reduce space and time requirements, this is not just a problem for wrapping. There are a few ways to solve this problem: 1.) Use partial specialization to leave these method definitions empty when not instantiating the proper dimension. This would probably be the correct solution if we could use it. 2.) Define full specializations of the method for all the wrapped types to leave the definition empty for the improper dimensions. This requires maintenance. 3.) Remove the methods altogether and require the general Rotate method to be used. I vote for this approach. -Brad From lorensen@crd.ge.com Wed Jan 28 20:12:34 2004 From: lorensen@crd.ge.com (Lorensen, William E (Research)) Date: Wed, 28 Jan 2004 15:12:34 -0500 Subject: [Insight-developers] RE: [Insight-users] TODO for AffineTransform::Rotate3D() Message-ID: Actually, I agree with 3). -----Original Message----- From: Brad King [mailto:brad.king@kitware.com] Sent: Wednesday, January 28, 2004 2:40 PM To: Luis Ibanez Cc: Lorensen, William E (Research); Insight Developers Subject: Re: [Insight-users] TODO for AffineTransform::Rotate3D() On Wed, 28 Jan 2004, Luis Ibanez wrote: > The concept checking for dimension seems to be ok when we only use C++, > since the methods are only compiled when used, but... wouldn't it > generate problems for wrapping ? We are instantiating this transform for > both 2D and 3D and the wrapping layer instantiates all the methods. They are only compiled when used if we are not doing explicit instantiation. An explicit instantiation will compile all methods. Since I'm looking into using explicit instantiation for common types to reduce space and time requirements, this is not just a problem for wrapping. There are a few ways to solve this problem: 1.) Use partial specialization to leave these method definitions empty when not instantiating the proper dimension. This would probably be the correct solution if we could use it. 2.) Define full specializations of the method for all the wrapped types to leave the definition empty for the improper dimensions. This requires maintenance. 3.) Remove the methods altogether and require the general Rotate method to be used. I vote for this approach. -Brad