pyushkevich/c3d

Convert3D tool

c3d's Issues

Failed to build on Ubuntu 20.04

c3d from the itk5 branch failed to build
Here is the biuld log: Biuld.log
Here is cmake cache.CMakeCache.txt

I am compiling on Ubuntu 20.04. I am using ITK 5.2.1. If you have any other questions please.

Building C3D on Ubuntu 20.04

I am trying to build C3D on Ubuntu 20.04 and I am using latest ITK version for it. But once I configure the build and try to make -j5 I get following error:

stefan@stefan-VirtualBox:~/c3d_test/c3d/build$ export CMAKE_PREFIX_PATH=$HOME/c3d_test/itk-c3d-install/usr/local
stefan@stefan-VirtualBox:~/c3d_test/c3d/build$ cmake ..
--    CPACK_SYSTEM_NAME Linux-x86_64
--    CPACK_PACKAGE_FILE_NAME c3d-1.1.0-Linux-x86_64
-- Configuring done
-- Generating done
-- Build files have been written to: /home/stefan/c3d_test/c3d/build
stefan@stefan-VirtualBox:~/c3d_test/c3d/build$ make -j5
Scanning dependencies of target ITKVoxBoIO
Scanning dependencies of target markdown_to_hex
Scanning dependencies of target cnd_maxflow
Scanning dependencies of target cnd_adapters
Scanning dependencies of target ITKPovRayIO
[  1%] Building C object CMakeFiles/markdown_to_hex.dir/utilities/hexdump.c.o
[  2%] Building CXX object CMakeFiles/cnd_maxflow.dir/external/GCv2p3/GCoptimization.cpp.o
[  3%] Building CXX object itkextras/VoxBoIO/CMakeFiles/ITKVoxBoIO.dir/itkVoxBoCUBImageIO.cxx.o
[  3%] Building CXX object itkextras/PovRayIO/CMakeFiles/ITKPovRayIO.dir/itkPovRayDF3ImageIO.cxx.o
[  4%] Linking C executable markdown_to_hex
[  4%] Built target markdown_to_hex
Scanning dependencies of target cnd_api
[  5%] Building CXX object CMakeFiles/cnd_api.dir/api/ConvertAPI.cxx.o
[  6%] Building CXX object CMakeFiles/cnd_maxflow.dir/external/GCv2p3/LinkedBlockList.cpp.o
[  6%] Building CXX object CMakeFiles/cnd_maxflow.dir/external/GCv2p3/graph.cpp.o
[  7%] Building CXX object CMakeFiles/cnd_maxflow.dir/external/GCv2p3/maxflow.cpp.o
[  8%] Building CXX object CMakeFiles/cnd_adapters.dir/adapters/AddImages.cxx.o
[  9%] Linking CXX static library libcnd_maxflow.a
[  9%] Built target cnd_maxflow
[ 10%] Building CXX object CMakeFiles/cnd_adapters.dir/adapters/AlignByLandmarks.cxx.o
[ 11%] Building CXX object itkextras/PovRayIO/CMakeFiles/ITKPovRayIO.dir/itkPovRayDF3ImageIOFactory.cxx.o
[ 11%] Linking CXX static library libcnd_api.a
[ 11%] Built target cnd_api
Scanning dependencies of target markdown_docs
[ 12%] Generating markdown_docs.h
[ 13%] Building CXX object itkextras/VoxBoIO/CMakeFiles/ITKVoxBoIO.dir/itkVoxBoCUBImageIOFactory.cxx.o
[ 13%] Built target markdown_docs
[ 13%] Building CXX object CMakeFiles/cnd_adapters.dir/adapters/AntiAliasImage.cxx.o
[ 14%] Linking CXX static library libITKPovRayIO.a
[ 14%] Built target ITKPovRayIO
Scanning dependencies of target cnd_driver
[ 15%] Building CXX object CMakeFiles/cnd_driver.dir/ConvertImageND.cxx.o
[ 16%] Linking CXX static library libITKVoxBoIO.a
[ 16%] Built target ITKVoxBoIO
[ 17%] Building CXX object CMakeFiles/cnd_driver.dir/utilities/doc/Documentation.cxx.o
[ 18%] Building CXX object CMakeFiles/cnd_adapters.dir/adapters/ApplyMetric.cxx.o
[ 19%] Building CXX object CMakeFiles/cnd_adapters.dir/adapters/BinaryHoleFill.cxx.o
/home/stefan/c3d_test/c3d/ConvertImageND.cxx: In member function ‘int ImageConverter<TPixel, VDim>::ProcessCommand(int, char**)’:
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:463:14: error: ‘vcl_acos’ was not declared in this scope
  463 |     adapter(&vcl_acos);
      |              ^~~~~~~~
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:516:14: error: ‘vcl_asin’ was not declared in this scope
  516 |     adapter(&vcl_asin);
      |              ^~~~~~~~
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:645:14: error: ‘vcl_cos’ was not declared in this scope
  645 |     adapter(&vcl_cos);
      |              ^~~~~~~
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:736:14: error: ‘vcl_exp’ was not declared in this scope; did you mean ‘ldexp’?
  736 |     adapter(&vcl_exp);
      |              ^~~~~~~
      |              ldexp
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:1005:14: error: ‘vcl_log’ was not declared in this scope
 1005 |     adapter(&vcl_log);
      |              ^~~~~~~
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:1012:14: error: ‘vcl_log10’ was not declared in this scope
 1012 |     adapter(&vcl_log10);
      |              ^~~~~~~~~
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx: In member function ‘double ApplyMetric<TPixel, VDim>::GetValueInternalSymmetric(ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::TransformPointer, ApplyMetric<TPixel, VDim>::TransformPointer, const char*)’:
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx:446:33: error: there are no arguments to ‘vcl_sqrt’ that depend on a template parameter, so a declaration of ‘vcl_sqrt’ must be available [-fpermissive]
  446 |     const double denom = -1.0 * vcl_sqrt(sff * smm );
      |                                 ^~~~~~~~
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx:446:33: note: (if you use ‘-fpermissive’, G++ will accept your code, but allowing the use of an undeclared name is deprecated)
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:1257:11: error: ‘vcl_exp’ was not declared in this scope; did you mean ‘ldexp’?
 1257 |     exp1(&vcl_exp);
      |           ^~~~~~~
      |           ldexp
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:1753:14: error: ‘vcl_sin’ was not declared in this scope
 1753 |     adapter(&vcl_sin);
      |              ^~~~~~~
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:1837:14: error: ‘vcl_sqrt’ was not declared in this scope
 1837 |     adapter(&vcl_sqrt);
      |              ^~~~~~~~
/home/stefan/c3d_test/c3d/ConvertImageND.cxx: In member function ‘int ImageConverter<TPixel, VDim>::ProcessCommandLine(int, char**)’:
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:2143:8: error: ‘itk::MultiThreader’ has not been declared
 2143 |   itk::MultiThreader::SetGlobalDefaultNumberOfThreads(1);
      |        ^~~~~~~~~~~~~
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx: In instantiation of ‘double ApplyMetric<TPixel, VDim>::GetValueInternalSymmetric(ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::TransformPointer, ApplyMetric<TPixel, VDim>::TransformPointer, const char*) [with TPixel = double; unsigned int VDim = 2; ApplyMetric<TPixel, VDim>::ImagePointer = itk::SmartPointer<itk::OrientedRASImage<double, 2> >; ApplyMetric<TPixel, VDim>::TransformPointer = itk::SmartPointer<itk::AffineTransform<double, 2> >]’:
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx:468:16:   required from here
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx:446:41: error: ‘vcl_sqrt’ was not declared in this scope
  446 |     const double denom = -1.0 * vcl_sqrt(sff * smm );
      |                                 ~~~~~~~~^~~~~~~~~~~~
/home/stefan/c3d_test/c3d/ConvertImageND.cxx: In instantiation of ‘TPixel ImageConverter<TPixel, VDim>::ReadIntensityValue(const char*) [with TPixel = double; unsigned int VDim = 2]’:
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:2959:16:   required from here
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:2386:28: error: ‘vnl_math_isnan’ was not declared in this scope; did you mean ‘vnl_math_h_’?
 2386 |         if (!vnl_math_isnan(*q))
      |              ~~~~~~~~~~~~~~^~~~
      |              vnl_math_h_
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx: In instantiation of ‘double ApplyMetric<TPixel, VDim>::GetValueInternalSymmetric(ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::TransformPointer, ApplyMetric<TPixel, VDim>::TransformPointer, const char*) [with TPixel = double; unsigned int VDim = 3; ApplyMetric<TPixel, VDim>::ImagePointer = itk::SmartPointer<itk::OrientedRASImage<double, 3> >; ApplyMetric<TPixel, VDim>::TransformPointer = itk::SmartPointer<itk::AffineTransform<double, 3> >]’:
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx:469:16:   required from here
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx:446:41: error: ‘vcl_sqrt’ was not declared in this scope
/home/stefan/c3d_test/c3d/ConvertImageND.cxx: In instantiation of ‘TPixel ImageConverter<TPixel, VDim>::ReadIntensityValue(const char*) [with TPixel = double; unsigned int VDim = 3]’:
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:2960:16:   required from here
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:2386:28: error: ‘vnl_math_isnan’ was not declared in this scope; did you mean ‘vnl_math_h_’?
 2386 |         if (!vnl_math_isnan(*q))
      |              ~~~~~~~~~~~~~~^~~~
      |              vnl_math_h_
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx: In instantiation of ‘double ApplyMetric<TPixel, VDim>::GetValueInternalSymmetric(ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::ImagePointer, ApplyMetric<TPixel, VDim>::TransformPointer, ApplyMetric<TPixel, VDim>::TransformPointer, const char*) [with TPixel = double; unsigned int VDim = 4; ApplyMetric<TPixel, VDim>::ImagePointer = itk::SmartPointer<itk::OrientedRASImage<double, 4> >; ApplyMetric<TPixel, VDim>::TransformPointer = itk::SmartPointer<itk::AffineTransform<double, 4> >]’:
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx:470:16:   required from here
/home/stefan/c3d_test/c3d/adapters/ApplyMetric.cxx:446:41: error: ‘vcl_sqrt’ was not declared in this scope
/home/stefan/c3d_test/c3d/ConvertImageND.cxx: In instantiation of ‘TPixel ImageConverter<TPixel, VDim>::ReadIntensityValue(const char*) [with TPixel = double; unsigned int VDim = 4]’:
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:2961:16:   required from here
/home/stefan/c3d_test/c3d/ConvertImageND.cxx:2386:28: error: ‘vnl_math_isnan’ was not declared in this scope; did you mean ‘vnl_math_h_’?
 2386 |         if (!vnl_math_isnan(*q))
      |              ~~~~~~~~~~~~~~^~~~
      |              vnl_math_h_
make[2]: *** [CMakeFiles/cnd_driver.dir/build.make:63: CMakeFiles/cnd_driver.dir/ConvertImageND.cxx.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:482: CMakeFiles/cnd_driver.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
[ 20%] Building CXX object CMakeFiles/cnd_adapters.dir/adapters/BiasFieldCorrectionN4.cxx.o
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx: In member function ‘void BiasFieldCorrectionN4<TPixel, VDim>::operator()()’:
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx:71:7: error: there are no arguments to ‘vcl_ceil’ that depend on a template parameter, so a declaration of ‘vcl_ceil’ must be available [-fpermissive]
   71 |       vcl_ceil( domain / splineDistance ) );
      |       ^~~~~~~~
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx:71:7: note: (if you use ‘-fpermissive’, G++ will accept your code, but allowing the use of an undeclared name is deprecated)
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx: In instantiation of ‘void BiasFieldCorrectionN4<TPixel, VDim>::operator()() [with TPixel = double; unsigned int VDim = 2]’:
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx:225:16:   required from here
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx:71:15: error: ‘vcl_ceil’ was not declared in this scope
   71 |       vcl_ceil( domain / splineDistance ) );
      |       ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx: In instantiation of ‘void BiasFieldCorrectionN4<TPixel, VDim>::operator()() [with TPixel = double; unsigned int VDim = 3]’:
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx:226:16:   required from here
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx:71:15: error: ‘vcl_ceil’ was not declared in this scope
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx: In instantiation of ‘void BiasFieldCorrectionN4<TPixel, VDim>::operator()() [with TPixel = double; unsigned int VDim = 4]’:
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx:227:16:   required from here
/home/stefan/c3d_test/c3d/adapters/BiasFieldCorrectionN4.cxx:71:15: error: ‘vcl_ceil’ was not declared in this scope
make[2]: *** [CMakeFiles/cnd_adapters.dir/build.make:102: CMakeFiles/cnd_adapters.dir/adapters/ApplyMetric.cxx.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[2]: *** [CMakeFiles/cnd_adapters.dir/build.make:128: CMakeFiles/cnd_adapters.dir/adapters/BiasFieldCorrectionN4.cxx.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:787: CMakeFiles/cnd_adapters.dir/all] Error 2
make: *** [Makefile:163: all] Error 2

Is this caused due to using wrong compiler or ITK version, or perhaps something else?

-pad-to issue

I'd like to use the -pad-to feature, but it's missing from the versions available on sourceforge (1.0.0 and 1.1.0); do I have to build from source to use this feature? Thanks.

Fails to build with a modern toolchain on conda-forge

Convert3D used to build on conda-forge but no longer does. I was trying to provide packages for the new osx-arm64 architecture (for the new Mac M-series), which requires a refresh of the feedstock. The old toolchain is no longer available.

conda-forge periodically updates their toolchain used for building packages, and a recent update broke the current build of Convert3D for version 1.3.0. Version 1.4.0 fails to build as well.

The logs from the Azure CI are available by following the GitHub workflow links down each PR discussion thread.

Misplaced parenthesis in -resample-mm

In ConvertImageND.cxx, the computation of sz is wrong due to a misplaced parenthesis.
It changes the dimension by one in case of small spacing.

The change is small:
sz[i] = static_cast<size_t>((0.5 + sz[i] * m_ImageStack.back()->GetSpacing()[i]) / vox[i]);
should be
sz[i] = static_cast<size_t>((0.5 + sz[i] * m_ImageStack.back()->GetSpacing()[i] / vox[i]));
I can create a PR if needed

-resample-iso raises the exception Unknown command

Hi, I was trying resampling options, it seems that c3d -resample-iso always raises the following exception:
Unknown command -resample-iso
Unknown exception caught by convert3d
When processing command: -resample-iso
Just wanted to report that
Bests

Enable processing 4d size vector for c4d

One of the size vector processing method, ReadRealVector, raising errors incorrectly for 4d size-vector, which causes some c4d command not working.

example: c4d -verbose img4d.nii.gz -resample-mm 2x2x2x2mm -o out.nii.gz

Provide conda-forge package

What's the statuts of (and work remaining for) porting Convert3D to ITK version 5?

I can see some initial work on the itk5 branch and a pending PR #8.

I'd be happy to help.

cli option to set value for SetGlobalDefaultCoordinateTolerance / SetGlobalDefaultDirectionTolerance

Currently the arguments to SetGlobalDefaultCoordinateTolerance and SetGlobalDefaultDirectionTolerance inside ConvertImageND.cxx is hardcoded to 1.0e-4, added in a38b52b

itk::ImageToImageFilterCommon::SetGlobalDefaultCoordinateTolerance(1.0e-4); itk::ImageToImageFilterCommon::SetGlobalDefaultDirectionTolerance(1.0e-4);

It would be great if this could also be set from the command line when needed.

I have the skills to help add this in a PR, but need some guide lines of how best to do this.

Offset in wrapped voxels with -wrap command

I'm trying to fix wraparound artifacts with -wrap, but the wrapped slices appear to be offset along the A-P axis.

Here's an example:

c3d blockWrapRPI.nii.gz -wrap 42x0x0 -o blockFixedRPI.nii.gz

blockWrapRPI.nii.gz
blockFixedRPI.nii.gz

This is on a Mac with c3d compiled from (e007137).

Compilation problem on Fedora

Trying to compile on Fedora 31, I get

...
$ make
Scanning dependencies of target markdown_to_hex
[  1%] Building C object CMakeFiles/markdown_to_hex.dir/utilities/hexdump.c.o
[  2%] Linking C executable markdown_to_hex
[  2%] Built target markdown_to_hex
Scanning dependencies of target markdown_docs
[  3%] Generating markdown_docs.h
[  3%] Built target markdown_docs
Scanning dependencies of target cnd_driver
[  4%] Building CXX object CMakeFiles/cnd_driver.dir/ConvertImageND.cxx.o
In file included from /usr/include/InsightToolkit/itkNumericTraits.h:51,
                 from /usr/include/InsightToolkit/itkConceptChecking.h:32,
                 from /usr/include/InsightToolkit/itkAtomicIntDetail.h:39,
                 from /usr/include/InsightToolkit/itkAtomicInt.h:38,
                 from /usr/include/InsightToolkit/itkTimeStamp.h:33,
                 from /usr/include/InsightToolkit/itkLightObject.h:23,
                 from /usr/include/InsightToolkit/itkObject.h:31,
                 from /usr/include/InsightToolkit/itkRegion.h:31,
                 from /usr/include/InsightToolkit/itkImageRegion.h:31,
                 from /usr/include/InsightToolkit/itkImage.h:21,
                 from /home/user/Downloads/gggg/c3d-git/itkextras/itkOrientedRASImage.h:4,
                 from /home/user/Downloads/gggg/c3d-git/ConvertImageND.h:29,
                 from /home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:26:
/usr/include/vxl/vcl/vcl_limits.h:11:5: warning: #warning "This header will be removed in future versions of VXL.  Use equivalent C++11 header instead. see: vxl/scripts/UseStandardHeaders.py" [-Wcpp]
   11 |   # warning "This header will be removed in future versions of VXL.  Use equivalent C++11 header instead. see: vxl/scripts/UseStandardHeaders.py"
      |     ^~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx: In member function ‘int ImageConverter::ProcessCommand(int, char**)’:
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:464:14: error: ‘vcl_acos’ was not declared in this scope
  464 |     adapter(&vcl_acos);
      |              ^~~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:517:14: error: ‘vcl_asin’ was not declared in this scope
  517 |     adapter(&vcl_asin);
      |              ^~~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:564:14: error: ‘vcl_ceil’ was not declared in this scope
  564 |     adapter(&vcl_ceil);
      |              ^~~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:653:14: error: ‘vcl_cos’ was not declared in this scope
  653 |     adapter(&vcl_cos);
      |              ^~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:744:14: error: ‘vcl_exp’ was not declared in this scope; did you mean ‘ldexp’?
  744 |     adapter(&vcl_exp);
      |              ^~~~~~~
      |              ldexp
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:800:14: error: ‘vcl_floor’ was not declared in this scope
  800 |     adapter(&vcl_floor);
      |              ^~~~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:1029:14: error: ‘vcl_log’ was not declared in this scope
 1029 |     adapter(&vcl_log);
      |              ^~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:1036:14: error: ‘vcl_log10’ was not declared in this scope
 1036 |     adapter(&vcl_log10);
      |              ^~~~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:1313:11: error: ‘vcl_exp’ was not declared in this scope; did you mean ‘ldexp’?
 1313 |     exp1(&vcl_exp);
      |           ^~~~~~~
      |           ldexp
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:1836:14: error: ‘vcl_sin’ was not declared in this scope
 1836 |     adapter(&vcl_sin);
      |              ^~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:1920:14: error: ‘vcl_sqrt’ was not declared in this scope
 1920 |     adapter(&vcl_sqrt);
      |              ^~~~~~~~
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx: In instantiation of ‘TPixel ImageConverter::ReadIntensityValue(const char*) [with TPixel = double; unsigned int VDim = 2]’:
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:3094:16:   required from here
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:2469:28: error: ‘vnl_math_isnan’ was not declared in this scope; did you mean ‘vnl_math_h_’?
 2469 |         if (!vnl_math_isnan(*q))
      |              ~~~~~~~~~~~~~~^~~~
      |              vnl_math_h_
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx: In instantiation of ‘TPixel ImageConverter::ReadIntensityValue(const char*) [with TPixel = double; unsigned int VDim = 3]’:
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:3095:16:   required from here
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:2469:28: error: ‘vnl_math_isnan’ was not declared in this scope; did you mean ‘vnl_math_h_’?
 2469 |         if (!vnl_math_isnan(*q))
      |              ~~~~~~~~~~~~~~^~~~
      |              vnl_math_h_
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx: In instantiation of ‘TPixel ImageConverter::ReadIntensityValue(const char*) [with TPixel = double; unsigned int VDim = 4]’:
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:3096:16:   required from here
/home/user/Downloads/gggg/c3d-git/ConvertImageND.cxx:2469:28: error: ‘vnl_math_isnan’ was not declared in this scope; did you mean ‘vnl_math_h_’?
 2469 |         if (!vnl_math_isnan(*q))
      |              ~~~~~~~~~~~~~~^~~~
      |              vnl_math_h_
make[2]: *** [CMakeFiles/cnd_driver.dir/build.make:63: CMakeFiles/cnd_driver.dir/ConvertImageND.cxx.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:329: CMakeFiles/cnd_driver.dir/all] Error 2

-rms doesn't compute RMS

% c3d -help rms                                                                                                                    
-rms                            : Voxelwise vector norm

Syntax: `-rms`

Computes RMS (root mean square) of all images on the stack. The command takes the square of each image on the stack, adds all the squared images and takes the square root of the result. This is very useful for statistical operations. Images must have the same size. 

    c3d img1.img img2.img img3.img img4.img -rms -o rms.img

The equivalent of this command is

    c3d img1.img img2.img img3.img img4.img -foreach -dup -times -endfor \
        -accum -add -endaccum -sqrt -o rms.img

In the code, it appears that -rms sets an option for an anti-aliasing filter. It doesn't actually compute the RMS.

-trans CLI option not recognised

The help dialog lists the translation flag as -trans but this is not recognised by the program, whereas -tran is

ITK-snap v3.8.0

fail to build c3d with the master branch

step to build:

git clone https://github.com/pyushkevich/c3d.git
cd c3d
cmake -B build -DBUILD_GUI=ON -DBUILD_TESTING=OFF -DCMAKE_CXX_FLAGS="-I/usr/include/eigen3" .
cd build
make

the commit I used is f7e9d5c
gcc: 11.2.0
itk: 5.2.1
eigen: 3.4.0
check the complete build log here

-origin not using NIFTI coordinates

c3d -help origin
-origin                         : Set image origin

Syntax: `-origin vector `

Set the origin of the image. The origin is the world coordinate (in NIfTI coordinate space) of the center of the voxel (0,0,0) in the image. The origin should be specified in millimeters. 

    c3d input.img -origin 100x100x100mm -o output.img

But the code calls SetOrigin

https://github.com/pyushkevich/c3d/blob/master/ConvertImageND.cxx#L1515-L1521

which is not overridden in itkOrientedRASImage.h .

The coordinate of voxel 0,0,0 should be the value of the qoffset field in the NIFTI header. But if I do

% c3d tpl-MNI152NLin2009cAsym_res-01_T1w.nii.gz -info-full | grep qoffset
    qoffset_x = -96
    qoffset_y = -132
    qoffset_z = -78
% c3d tpl-MNI152NLin2009cAsym_res-01_T1w.nii.gz -origin -96x-132x-78mm -o origin_changed.nii.gz

the origin is changed

% c3d tpl-MNI152NLin2009cAsym_res-01_T1w.nii.gz -origin -96x-132x-78mm -info-full  | grep qoffset
    qoffset_x = 96
    qoffset_y = 132
    qoffset_z = -78