ITK/Release 4/Modularization/Prototype/Partition: Difference between revisions

From KitwarePublic
Jump to navigationJump to search
No edit summary
Line 4: Line 4:
* common (Object, Exception,ProcessObject ...)
* common (Object, Exception,ProcessObject ...)
* io-common
* io-common
* itksys
* itkvnl


= Numerics =
= Numerics =
Line 55: Line 57:
* Thresholding
* Thresholding
* Lable Voting
* Lable Voting
* Fast marching




Line 68: Line 71:


= Others =
= Others =
(Utility modules)

Revision as of 22:37, 6 November 2010

The following is the suggested list of modules in which ITK will be partitioned. Those modules are conceptually categorized into Core, IO, Filtering Segmentation, Registration and Others, according to their functionality.

Core

  • common (Object, Exception,ProcessObject ...)
  • io-common
  • itksys
  • itkvnl

Numerics

  • Optimizers
  • Statistics
  • Neural Networks
  • FEM

IO

One module per file format. This is because every file format brings in a dependency on a full third party library

  • PNG
  • JPEG
  • Nrrd
  • GE
  • Analyze
  • BMP
  • Gipl
  • Nifti
  • DICOM
  • Meta
  • Siemens
  • XML
  • Transform

Filtering

  • Anisotropic Diffusion
  • Isotropic (Mean, Median, Gaussian)
  • Thresholding (BinaryThreshold, OtsuThreshold)
  • Mathematical Morphology
  • Hole Filling
  • Anti-alias
  • Gradient
  • Thinning
  • Pruning
  • Ballon Force
  • Curvature Flow
  • Gradient vector flow

Segmentation

  • Region Growing
  • Level Sets
  • Deformable Mesh
  • Watersheds
  • Voronoi
  • Kmeans
  • Classifier
  • MRF
  • GibbsPiror
  • ConnectedComponent
  • Thresholding
  • Lable Voting
  • Fast marching


Registration

  • registration-common (common classes for all image registration methods, eg. Deformation Fields)
  • Deformable Registration
  • Demons
  • BSpline
  • FEM Registration
  • Curvature Registration
  • Image to Spatial Object registration
  • Point set to point set registration

Others

(Utility modules)