ITK 4.7.0 has been released!

 

On behalf of the Insight Toolkit community, we are proud to announce that ITK 4.7.0 has been released!

Get it now!

 

The 4.7 release is a major milestone that marks the hard work of many outstanding community members. Congratulations and well done to the 29 contributors to this release. We would especially like to recognize the new contributors, Gert Wollny, Girish Mallya, Christopher Mullins, Umang B, David T. Chen, Danny Perry, Johan Andruejol, and Ismael Belghiti.

 

 

Among the major contributions in this release: all ITK examples from the Software Guide, Wiki Examples, and SphinxExamples are new indexed and crossed linked in Doxygen — when you visit the Doxygen page for your favorite filter or other object, say the ResampleImageFilter, there is a large selection of examples in the Examples section. TIFF IO has undergone a significant cleanup and speed-up, and TIFF tags are now read into the Image’s MetaDataDictionary. Three new Remote Modules are available, for gradient computation, DICOM IO, and QuadEdgeMesh filtering. Python wrapping, CMake configuration, and registration continue to improve. A new, experimental thread pool is available — to enable set the ITK_USE_THREADPOOL or call itk::MultiThreader::SetGlobalDefaultUseThreadPool(true).

 

Thanks to everyone for their hard work!



Enjoy ITK!

 

 

New Features

 

* Documentation Improvements

 – Update Software Guide registration examples to ITKv4 framework

 – Wiki and Sphinx examples are indexed in Doxygen

 – Clean Software Guide dashboard build

 

* Filtering Improvements

 – New Laplacian deformation filter’s for QuadEdgeMesh’s

   – IJ article: http://hdl.handle.net/10380/3410

– FFTComplexToComplexImageFilter out of Review, Vnl, FFTW implementations

– Faster recursive Gaussian on VectorImage

 

* ImageIO improvements

 – PNG sCAL unit support

 – TIFF IO performance greatly improved

 – TIFF tags read into Image MetaDataDictionary

 – MetaIO updated

 – Better support for system GDCM

 

* Infrastructure improvements

 – New method to erase a MetaDataDictionary entry

 – New global method to set physical space tolerance

 – New, experimental thread pool support

 – NumericTraits::Zero and ::One are deprecated in favor of ::ZeroValue() and ::OneValue()

 – VTK bridge modules only request required VTK modules

 

* New Remote Modules

 – Higher Order Accurate Gradient

   – http://hdl.handle.net/10380/3231

 – IO Transform DCMTK

   – http://hdl.handle.net/10380/3468

 – Subdivision Quad Edge Mesh Filters

   – http://hdl.handle.net/10380/3307

 

* Registration improvements

 – LandmarkBasedTransformInitializer supports BSplineTransform

 – Transverse inverse consistency increased

 – Automatic initialization of the transform center for multi-stage registrations

 – Sparse point B-spline field estimate

 – Mutual information faster, lower memory usage

 

* Wrapping improvements

 – Wrapping now builds without type warnings

 – pygccxml bumped to 1.6.2

 – Latest GCCXML

 – CMake wrapping infrastructure cleanup

 – Build with Ninja CMake generator

 – New GetTypes() and GetTypesAsList() methods

 – ITK_WRAP_PYTHON is exposed in ITKConfig.cmake

 

* Many style improvements — ITK gets more stylish with every release!

 

* Improved code coverage — we are over 85%!

 

* *Lots* of important bug fixes

* And much more! See details in the release changelog.

 

 

Questions or comments are always welcome!