[Gate-users] Fixed Forced Detection Actor
Triltsch, Nicolas
nicolas.triltsch at tum.de
Wed Nov 30 14:36:30 CET 2016
Hey,
This is what I found in my CMakeCache.txt file:
//The directory containing RTKConfig.cmake. This is either the
// root of the build tree, or PREFIX/lib for an installation.
RTK_DIR:PATH=/usr/local/bin/RTK-1.3.0-build
Which should be correct!
In the first installation of ITK and RTK I didn't change in ccmake the
option CMAKE_INSTALL_PREFIX. The second time I compiled ITK and RTK, I
changed CMAKE_INSTALL_PREFIX=/usr/local/bin/RTK-1.3.0-install (a
installation folder I created). Additionally I created I folder
RTK-1.3.0-build for starting ccmake ../RTK-1.3.0 and compiling the
source code.
Do you think there could be the error?
Nico
On 11/30/2016 01:35 PM, Simon Rit wrote:
> I don't know what's going on but the fact that you have to change
> LD_LIBRARY_PATH is very suspicious. I would check in the
> CMakeCache.txt of your Gate compilation directory what is the value of
> RTK_DIR. It seems to me that you might have mixed several versions of
> RTK.
> Simon
>
> On Wed, Nov 30, 2016 at 1:30 PM, Triltsch, Nicolas
> <nicolas.triltsch at tum.de> wrote:
>> Hey David and Simon,
>>
>> The libRTK in my case is located
>> /usr/local/bin/RTK-1.3.0-install/lib/libRTK.so . If I set the
>> LD_LIBRARY_PATH, Gate is starting. But after do a testrun with
>> FixedForcedDetection Actor, it states the following:
>> "===========================================================
>> There was a crash.
>> This is the entire stack trace of all threads:
>> ===========================================================
>> #0 0x00007f096571751b in __GI___waitpid (pid=20648, stat_loc=stat_loc
>> entry=0x7ffdb4477b40, options=options
>> entry=0) at ../sysdeps/unix/sysv/linux/waitpid.c:29
>> #1 0x00007f0965690fbb in do_system (line=<optimized out>) at
>> ../sysdeps/posix/system.c:148
>> #2 0x00007f096d69ddd4 in TUnixSystem::StackTrace() () from
>> /usr/lib/x86_64-linux-gnu/libCore.so.5.34
>> #3 0x00007f096d6a003c in TUnixSystem::DispatchSignals(ESignals) () from
>> /usr/lib/x86_64-linux-gnu/libCore.so.5.34
>> #4 <signal handler called>
>> #5 0x000000000353d050 in ?? ()
>> #6 0x00007f096b20b582 in itk::(anonymous
>> namespace)::CleanUpObjectFactory::~CleanUpObjectFactory() () from
>> /usr/local/bin/RTK-1.3.0-install/lib/libRTK.so
>> #7 0x00007f096568636a in __cxa_finalize (d=0x7f096bd2ec40) at
>> cxa_finalize.c:56
>> #8 0x00007f096b158e43 in __do_global_dtors_aux () from
>> /usr/local/bin/RTK-1.3.0-install/lib/libRTK.so
>> #9 0x00007ffdb447a810 in ?? ()
>> #10 0x00007f096f2dac17 in _dl_fini () at dl-fini.c:235
>> ==========================================================="
>>
>> Maybe I should compile RTK again and save the library under /usr/lib?
>> Nico
>>
>>
>> On 11/30/2016 01:18 PM, Simon Rit wrote:
>>> Hi,
>>> I agree with everything pointed out by David. I have checked that
>>> compiling RTK with dynamic libraries did work and it does so I'm not
>>> sure what has happened on your side but it should find libRTK if you
>>> haven't moved it after compilation.
>>> Simon
>>>
>>> On Wed, Nov 30, 2016 at 11:59 AM, David Boersma
>>> <david.boersma at physics.uu.se> wrote:
>>>> Hi Nicolas,
>>>>
>>>> Den 30/11/2016 kl. 11:48, skrev Triltsch, Nicolas:
>>>>> Hey Simon,
>>>>>
>>>>> I compiled everything again and there were no more errors:) But if I
>>>>> start Gate it states the following:
>>>>>
>>>>> /"Gate: error while loading shared libraries: libRTK.so: cannot open
>>>>> shared object file: No such file or directory"/
>>>>>
>>>>> Why does it not find the libraries?
>>>>
>>>> Where is libRTK.so installed on your system?
>>>>
>>>> If it is in some standard directory, like /usr/lib, then it should just
>>>> be
>>>> found without any extra actions. So I assume that you installed it
>>>> somewhere
>>>> else. Do not move libraries around after installing (because sometimes
>>>> their
>>>> paths are hardcoded inside the libraries and other code).
>>>>
>>>> Let's say the libRTK.so is installed in /A/B/C/lib, then you could try
>>>>
>>>> export LD_LIBRARY_PATH=/A/B/C/lib:$LD_LIBRARY_PATH
>>>>
>>>> Then try running Gate again. Or run "ldd Gate" to inspect the list of
>>>> libraries that the runtime linker is finding (or not finding) for it.
>>>>
>>>> (If you are on MacOSX, then you need to set DYLD_LIBRARY_PATH instead,
>>>> and
>>>> instead of "ldd" you run "otool -L".)
>>>>
>>>> HTH,
>>>> /DavidB
>>>>
>>>>
>>>>> Cheers, Nico
>>>>>
>>>>> On 11/29/2016 10:09 PM, Simon Rit wrote:
>>>>>> Hi,
>>>>>> I'm not sure what happened. Did you maybe update ITK without
>>>>>> recompiling it? Anyway, I'd suggest to recompile ITK, RTK and Gate (in
>>>>>> this order). Sorry for the trouble, I know these dependencies are a
>>>>>> bit inconvenient.
>>>>>> Simon
>>>>>>
>>>>>> On Tue, Nov 29, 2016 at 2:06 PM, Triltsch, Nicolas
>>>>>> <nicolas.triltsch at tum.de <mailto:nicolas.triltsch at tum.de>> wrote:
>>>>>>
>>>>>> Hey Simon,
>>>>>>
>>>>>> I followed your two changes 1) and 3) and after that I compiled
>>>>>> Gate again, in order to make the changes valid. But now I get an
>>>>>> error message when compiling gate_v7.2, which says:
>>>>>>
>>>>>> "/
>>>>>> /
>>>>>>
>>>>>> /make[2]: *** No rule to make target
>>>>>>
>>>>>>
>>>>>> '/usr/local/bin/InsightToolkit-4.10.1/InsightToolkit-4.10.1-build/lib/libITKReview-4.10.so.1',
>>>>>> needed by 'Gate'. Stop.//
>>>>>> //make[2]: *** Waiting for unfinished jobs....//
>>>>>> //[100%] Building CXX object
>>>>>>
>>>>>>
>>>>>> CMakeFiles/Gate.dir/source/externals/clhep/src/CLHEP/RandomObjects/RandomVector.cc.o//
>>>>>> //CMakeFiles/Makefile2:67: recipe for target
>>>>>> 'CMakeFiles/Gate.dir/all' failed//
>>>>>> //make[1]: *** [CMakeFiles/Gate.dir/all] Error 2//
>>>>>> //Makefile:127: recipe for target 'all' failed//
>>>>>> //make: *** [all] Error 2/
>>>>>>
>>>>>> "
>>>>>>
>>>>>> Honestly, I have no idea what to do. I think it has something to
>>>>>> do with the ITK package, but I don't know why. If you have an idea
>>>>>> that would be great, as my Gate is not running any more:/ If I
>>>>>> type 'Gate' in the terminal, it says now:
>>>>>>
>>>>>> "
>>>>>>
>>>>>> /Gate: error while loading shared libraries:
>>>>>> libITKIOBMP-4.10.so.1: cannot open shared object file: No such
>>>>>> file or directory/
>>>>>>
>>>>>> "
>>>>>>
>>>>>> Thanks in advance!
>>>>>>
>>>>>> Nico
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 11/28/2016 03:57 PM, Simon Rit wrote:
>>>>>>> Hi,
>>>>>>>
>>>>>>> On Mon, Nov 28, 2016 at 2:00 PM, Triltsch, Nicolas
>>>>>>> <nicolas.triltsch at tum.de <mailto:nicolas.triltsch at tum.de>> wrote:
>>>>>>>
>>>>>>> Hey everyone,
>>>>>>>
>>>>>>> I have three questions concerning the fixed forced detection
>>>>>>> actor(ffda):
>>>>>>>
>>>>>>> 1) As output one can generate e.g. primary.mha, compton.mha,
>>>>>>> rayleigh.mha and total.mha images. In the GATE wiki it
>>>>>>> states, that primary + compton + rayleigh = total. But if I
>>>>>>> compare the two images they are different! Why? (total.mha
>>>>>>> image was flatfield corrected by default; primary, compton
>>>>>>> and rayleigh weren't, so I used the flatField.mha to correct
>>>>>>> them before the comparison with total.mha)
>>>>>>>
>>>>>>> I checked and there is a bug in the total filename. However, it
>>>>>>> does not use the flat field but rather one step was forgotten
>>>>>>> (accounting for the total number of particles in the simulation).
>>>>>>> I must admit that I never use that output in practice, hence the
>>>>>>> bug. This fix
>>>>>>>
>>>>>>>
>>>>>>> <https://github.com/OpenGATE/Gate/commit/988f886133ee741eeca692df027e7786e025c7bc>
>>>>>>> should correct that.
>>>>>>>
>>>>>>>
>>>>>>> 2) For the ffda the source needs to be focused. Why is that
>>>>>>> necessary?
>>>>>>>
>>>>>>> Because ffda uses RTK <http://openrtk.org/> for deterministic
>>>>>>> computation and RTK is working for point sources.
>>>>>>>
>>>>>>>
>>>>>>> 3) One output image is called secondary.mha. First, I thought
>>>>>>> this image would represent secondary scatter events. But in
>>>>>>> the GATE wiki it says that secondary = compton + rayleigh. Is
>>>>>>> their a way to obtain a secondary scatter (multiple scatter)
>>>>>>> image?
>>>>>>>
>>>>>>> It used to be possible but I think we have removed it to
>>>>>>> "minimize" the number of options. You can try (but I don't
>>>>>>> guarantee it works) to add a line in the code, e.g. around line
>>>>>>> 57 of GateFixedForcedDetectionActor.cc:
>>>>>>> mPerOrderImagesBaseName="order";
>>>>>>> It will then create an image of the form
>>>>>>> "orderRayleigh0004_order02.mha" for the second order of the
>>>>>>> rayleigh projection number 4.
>>>>>>> Simon
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> I would appreciate any help!
>>>>>>>
>>>>>>> Nico
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Gate-users mailing list
>>>>>>> Gate-users at lists.opengatecollaboration.org
>>>>>>> <mailto:Gate-users at lists.opengatecollaboration.org>
>>>>>>>
>>>>>>> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users
>>>>>>>
>>>>>>> <http://lists.opengatecollaboration.org/mailman/listinfo/gate-users>
>>>>>>>
>>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Gate-users mailing list
>>>>> Gate-users at lists.opengatecollaboration.org
>>>>> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users
>>>>>
>>>> _______________________________________________
>>>> Gate-users mailing list
>>>> Gate-users at lists.opengatecollaboration.org
>>>> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users
>>> _______________________________________________
>>> Gate-users mailing list
>>> Gate-users at lists.opengatecollaboration.org
>>> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users
>>
More information about the Gate-users
mailing list