ITK Release 4/SimpleITK/Tcon 2012 2 29

From KitwarePublic
< ITK Release 4‎ | SimpleITK
Revision as of 08:49, 29 February 2012 by (talk) (→‎Daily Details)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

Tcon Information

  Please join the following teleconference line:
      Toll-Free #:      1-888-677-0429
      Participant Code: 96280
  Wednesday, February 29th, 9:30am Eastern Time


  • Bradly Lowekamp
  • Dan Blezek


Release 0.4.0

  • Currently 82 binary downloads
  • Patch 0.4.1

Daily Details

  • The Stage Repository
            SIMPLEITK-248_UpdateITKTo41 | master=0 next=1
      SIMPLEITK-251_BreakupBasicFilters | master=0 next=1
       SIMPLEITK-259_ApplyApacheLicense | master=0 next=1
SIMPLEITK-224_AddVectorSupportGetBuffer | master=0 next=0
   SIMPLEITK-256_SetCompilerForSwigPCRE | master=0 next=1
         SIMPLEITK-98_add_install_rules | master=0 next=0
SIMPLEITK-3_RecursiveGaussianForVectors | master=0 next=1
   SIMPLEITK-211_AddingWatershedFilters | master=0 next=0
SIMPLEITK-244-PackageSimpleITKForPython | master=0 next=1
 SIMPLEITK-98_add_install_rules-renewed | master=0 next=0
       SIMPLEITK-1_AddImageRegistration | master=0 next=0
SIMPLEITK-255-PackageSimpleITKForPython | master=0 next=1
                    UpdateDataSubmodule | master=0 next=1

Outstanding Test Failures

Priorities for next release

  • Registration
  • C++ installation
  • Splitting Wrapping directory into languages
  • Splitting BasicFilters Library
  • improved vector image support

Other Open Topics

  • Wiki and Doxygen pages should be review for release.
  • Adding support for import and export vector images to numpy
  • Registration
    • to be major new feature to work on in Febuary
  • C++ installation

R Wrapping Follow-up

  • working on fixing the automatic testing for R
  • Richard Beare is hacking away he is reporting the following:
I'm thinking about import options based on my old R imaging package.
Images were also represented by a pointer to a C object (external
reference). By providing the right generic operators we could treat
the image as an array for purposes of subsetting. This meant that you
could do something like this:

im2 <- im1[100:200,]  # im2 is another image, not an array, with all
the columns and rows 100 to 200

If we wanted an array version of an image we'd do
im2 <- as.array(im1)

as.array is a generic R operation that we specialised for images.

The SimpleITK bindings are already capable of doing:

pixval <- im1[c(100,200)]

so we're well on the way.

The other neat thing I'd like is operator overloading. R has
facilities for that, so that you should be able to write expressions
which treat images just like any other variable.1


  • Discuss long term goals and milestones.
  • Documentations
  • Automate Script to List filters wrapped
    • This script should take as input a list of ITK Modules or Directories, for each modual determine which filters are implemented in SimpleITK
    • Additional input would include filter/files to exclude
    • Output could be a dox file, so that it could be a doxygen related page
  • Registration is very important