|
| class | npl::RigidInfoComp |
| | The Rigid MI Computer is used to compute the mutual information and gradient of mutual information between two images. As the name implies, it is designed for 6 parameter rigid transforms. More...
|
| |
| class | npl::RigidCorrComp |
| | The Rigid Corr Computer is used to compute the correlation and gradient of correlation between two images. As the name implies, it is designed for 6 parameter rigid transforms. More...
|
| |
| class | npl::ProbDistCorrInfoComp |
| | The distortion correction MI Computer is used to compute the mutual information and gradient of mutual information between two images using nonrigid, unidirectional, B-spline transform. In this variant the moving image should be a probability map. More...
|
| |
| class | npl::DistCorrInfoComp |
| | The distortion correction MI Computer is used to compute the mutual information and gradient of mutual information between two images using nonrigid, unidirectional, B-spline transform. More...
|
| |
| struct | npl::Rigid3DTrans |
| | Struct for holding information about a rigid transform. Note that rotation R = Rx*Ry*Rz, where Rx, Ry, and Rz are the rotations about x, y and z aaxes, and the angles are stored (in radians) in the rotation member. More...
|
| |
|
| ptr< MRImage > | npl::motionCorrect (ptr< const MRImage > input, size_t ref) |
| | Performs motion correction on a set of volumes. Each 3D volume is extracted and linearly registered with the ref volume. More...
|
| |
| Rigid3DTrans | npl::corReg3D (ptr< const MRImage > fixed, ptr< const MRImage > moving, const std::vector< double > &sigmas) |
| | Performs correlation based registration between two 3D volumes. note that the two volumes should have identical sampling and identical orientation. If that is not the case, an exception will be thrown. More...
|
| |
| Rigid3DTrans | npl::informationReg3D (ptr< const MRImage > fixed, ptr< const MRImage > moving, const std::vector< double > &sigmas, size_t nbins=128, size_t binradius=4, std::string metric="MI", double stopx=0.001) |
| | Performs information-based registration between two 3D volumes. note that the two volumes should have identical sampling and identical orientation. If that is not the case, an exception will be thrown. More...
|
| |
| ptr< MRImage > | npl::infoDistCor (ptr< const MRImage > fixed, ptr< const MRImage > moving, bool otsu, int dir, double bspace, double jac, double tps, const std::vector< double > &sigmas, size_t nbins=128, size_t binradius=4, string metric="MI", size_t hist=8, double stopx=1e-5, double beta=0.5) |
| | Information based registration between two 3D volumes. note that the two volumes should have identical sampling and identical orientation. More...
|
| |
| int | npl::cor3DDerivTest (double step, double tol, ptr< const MRImage > in1, ptr< const MRImage > in2) |
| | This function checks the validity of the derivative functions used to optimize between-image corrlation. More...
|
| |
| int | npl::information3DDerivTest (double step, double tol, ptr< const MRImage > in1, ptr< const MRImage > in2) |
| | This function checks the validity of the derivative functions used to optimize between-image corrlation. More...
|
| |
| int | npl::distcorDerivTest (double step, double tol, shared_ptr< const MRImage > in1, shared_ptr< const MRImage > in2, double regj=0, double regt=0) |
| | This function checks the validity of the derivative functions used to optimize between-image corrlation. More...
|
| |
| std::ostream & | npl::operator<< (std::ostream &stream, const Rigid3DTrans &rigid) |
| | Prints a rigid transform. More...
|
| |
Registration functions are implemented with classes linked to optimization functions. All registration algorithms ultimately may be performed with a simple function call, but the Computer classes (of which there is currently just RigidCorrComp) are exposed in case you want to use them for your own registration algorithms.
A computer is needed for every pair of Metric and Transform type.
Note that for rigid computers the state_x variable will store the final result. The first 3 parameters are the rotation about the X, Y, Z axes, in degrees. The second 3 will be the shift in X, Y, Z in mm.
| int npl::cor3DDerivTest |
( |
double |
step, |
|
|
double |
tol, |
|
|
ptr< const MRImage > |
in1, |
|
|
ptr< const MRImage > |
in2 |
|
) |
| |
This function checks the validity of the derivative functions used to optimize between-image corrlation.
- Parameters
-
| step | Test step size |
| tol | Tolerance in error between analytical and Numeric gratient |
| in1 | Image 1 |
| in2 | Image 2 |
- Returns
- 0 if success, -1 if failure
| Rigid3DTrans npl::corReg3D |
( |
ptr< const MRImage > |
fixed, |
|
|
ptr< const MRImage > |
moving, |
|
|
const std::vector< double > & |
sigmas |
|
) |
| |
Performs correlation based registration between two 3D volumes. note that the two volumes should have identical sampling and identical orientation. If that is not the case, an exception will be thrown.
- Todo:
- make it v = Ru + s, then u = INV(R)*(v - s)
- Parameters
-
| fixed | Image which will be the target of registration. |
| moving | Image which will be rotated then shifted to match fixed. |
| sigmas | Standard deviation of smoothing at each level |
- Returns
- Output rigid transform
| int npl::distcorDerivTest |
( |
double |
step, |
|
|
double |
tol, |
|
|
shared_ptr< const MRImage > |
in1, |
|
|
shared_ptr< const MRImage > |
in2, |
|
|
double |
regj = 0, |
|
|
double |
regt = 0 |
|
) |
| |
This function checks the validity of the derivative functions used to optimize between-image corrlation.
- Parameters
-
| step | Test step size |
| tol | Tolerance in error between analytical and Numeric gratient |
| in1 | Image 1 |
| in2 | Image 2 |
| regj | Jackobian weight |
| regt | Thin-plate-spline weight |
- Returns
- 0 if success, -1 if failure
| ptr<MRImage> npl::infoDistCor |
( |
ptr< const MRImage > |
fixed, |
|
|
ptr< const MRImage > |
moving, |
|
|
bool |
otsu, |
|
|
int |
dir, |
|
|
double |
bspace, |
|
|
double |
jac, |
|
|
double |
tps, |
|
|
const std::vector< double > & |
sigmas, |
|
|
size_t |
nbins = 128, |
|
|
size_t |
binradius = 4, |
|
|
string |
metric = "MI", |
|
|
size_t |
hist = 8, |
|
|
double |
stopx = 1e-5, |
|
|
double |
beta = 0.5 |
|
) |
| |
Information based registration between two 3D volumes. note that the two volumes should have identical sampling and identical orientation.
- Parameters
-
| fixed | Image which will be the target of registration. |
| moving | Image which will be rotated then shifted to match fixed. |
| otsu | perform OTSU thresholding on images after downsampling |
| dir | direction/dimension of distortion |
| bspace | Spacing of B-Spline knots (in mm) |
| jac | jacobian regularizer weight |
| tps | thin-plate-spline regularizer weight |
| sigmas | Standard deviation of smoothing at each level |
| nbins | Number of bins in marginal PDF |
| binradius | radius of parzen window, to smooth pdf |
| metric | Type of information based metric to use |
| hist | length of history to keep |
| stopx | Minimum step size before stopping |
| beta | Fraction of previous step size to consider for next step size. Essentially this decides how quickly to reduce step sizes (note 1 would mean NO reduction and would continue forever, 0 would do 1 step then stop) |
- Returns
- parameters of bspline
| int npl::information3DDerivTest |
( |
double |
step, |
|
|
double |
tol, |
|
|
ptr< const MRImage > |
in1, |
|
|
ptr< const MRImage > |
in2 |
|
) |
| |
This function checks the validity of the derivative functions used to optimize between-image corrlation.
- Parameters
-
| step | Test step size |
| tol | Tolerance in error between analytical and Numeric gratient |
| in1 | Image 1 |
| in2 | Image 2 |
- Returns
- 0 if success, -1 if failure
| Rigid3DTrans npl::informationReg3D |
( |
ptr< const MRImage > |
fixed, |
|
|
ptr< const MRImage > |
moving, |
|
|
const std::vector< double > & |
sigmas, |
|
|
size_t |
nbins = 128, |
|
|
size_t |
binradius = 4, |
|
|
std::string |
metric = "MI", |
|
|
double |
stopx = 0.001 |
|
) |
| |
Performs information-based registration between two 3D volumes. note that the two volumes should have identical sampling and identical orientation. If that is not the case, an exception will be thrown.
- Parameters
-
| fixed | Image which will be the target of registration. |
| moving | Image which will be rotated then shifted to match fixed. |
| sigmas | Standard deviation of smoothing kernel at each level |
| nbins | Number of bins during marginal density estimation (joint with have nbins*nbins) |
| binradius | During parzen window, the radius of the smoothing kernel |
| metric | metric to use (default is MI) |
| stopx | Smallest step to take, stop after steps reach this size |
- Returns
- Rigid transform.
| ptr<MRImage> npl::motionCorrect |
( |
ptr< const MRImage > |
input, |
|
|
size_t |
ref |
|
) |
| |
Performs motion correction on a set of volumes. Each 3D volume is extracted and linearly registered with the ref volume.
- Parameters
-
| input | 3+D volume (set of 3D volumes, all higher dimensions are treated equally as separate volumes). |
| ref | Reference t, all images will be registered to the specified timepoint |
- Returns
- Motion corrected volume.
| std::ostream& npl::operator<< |
( |
std::ostream & |
stream, |
|
|
const Rigid3DTrans & |
rigid |
|
) |
| |
|
inline |
Prints a rigid transform.
- Parameters
-
| stream | Output stream |
| rigid | Rigid transform |
- Returns
- after this is inserted, stream
Definition at line 1088 of file registration.h.
1.8.9.1 
