Affine Transform CLP

From user's Wiki!
Jump to: navigation, search
Affine Transform

General Information

  • Type: CLP
  • Category: Registration
  • Author: Tristan Whitmarsh
  • Contributor: Xavi Planes
  • Contact: xavier.planes at upf.edu
  • Description: Compute an affine transformation between two images. The algorithm uses Mattes’ implementation of the mutual information similarity metric [3], and employs the simplex method for optimization.

[3] Mattes, Haynor, Vesselle, Lewellen, Eubank. “PET-CT image registration in the chest using free-form deformations,” IEEE Transactions on Medical Imaging 22(1):120-128, January 2003. Doi: 10.1109/TMI.2003.809072)

Usage

Parameters

Label Type Flag / Index Default value / Channel Description
Pre processing
Region region region 0,0,0,0,0,0 user specified region for fixed image
Sigma double sigma_smooth 0 Sigma for the Gaussian filter
Low Thresh Fixed double flaglowThreshFixed false Low Thresh Fixed
lowThreshFixed float lowThreshFixed 0 Pixels below this intensity are ignored for metric computation
High Thresh Fixed float flaghighThreshFixed false High Thresh Fixed
highThreshFixed float highThreshFixed 0 Pixels above this intensity are ignored for metric computation
Low Thresh Moving boolean flaglowThreshMoving false Low Thresh Moving
lowThreshMoving float lowThreshMoving 0 Pixels below this intensity are ignored for metric computation
High Thresh Moving boolean flaghighThreshMoving false High Thresh Moving
highThreshMoving float highThreshMoving 0 Pixels above this intensity are ignored for metric computation
Registration
initializeTrans boolean initializeTrans false Initialize the transform using CenteredTransformInitializer
Num of iterations integer nb_Itr 100 Num of iterations
nb_maxSteplng float nb_maxSteplng 0.5 Maximum step length> The initial step length
nb_minSteplng float nb_minSteplng 0.0001 Minimum step length> The tolerance for convergence
Num of bins integer nb_Bins 64 Number of histogram bins
Fraction of samples integer frac_S 100 Fraction of samples
paddingValue float paddingValue 0 Default pixel value
add_observer boolean add_observer true show registration outputs
IO
Fixed Image image index 0 input Fixed Image
Moving Image image index 1 input Moving Image
Mask Image image maskFileName input Mask Image
Output Image image outputImage output Output Image
Output transform transform outputtransform output Transform calculated that aligns the fixed and moving image. Maps positions from the fixed coordinate frame to the moving coordinate frame. Optional (specify an output transform or an output volume or both)

Examples

Affine Transform between two images. The last column is the result and the transformation grid.

Affine Transform

Development

Source code: C++ Source code and XML description

Go back to Users