@ALDParametrizedClass public abstract class MTBLengthEnergyFW2DNonPDE extends MTBLengthEnergyFWNonPDE
doNormalization, FLAG_LENGTH, mu, scaleFactordebug, name| Constructor and Description |
|---|
MTBLengthEnergyFW2DNonPDE()
constructor
|
MTBLengthEnergyFW2DNonPDE(double mu)
constructor
|
| Modifier and Type | Method and Description |
|---|---|
double |
deltaE(int x,
int y,
int z,
short newPhase,
MTBLevelsetMembership phi)
Calculate differences of energy if the pixel/voxel (x,y,z) is changed
from its current phase in the level set function
phi supplied
from current phase to newPhase. |
double |
E(int x,
int y,
MTBLevelsetMembership phi)
Get energy term a pixel (x,y) for
phi. 2D only |
double |
E(MTBLevelsetMembership phi)
Return the complete energy for the level set function
phi. |
protected abstract double |
gradMagHeaviside(double a,
double b,
double c)
calculate the gradient magnitude of the foreward difference of the heavyside function
corresponding to the level set for one pixel
|
createEnergy, initsetScaleFactor, toStringdeltaE, estimateParams, getNewIndent, print, setDebug, updateParams, updateParamspublic MTBLengthEnergyFW2DNonPDE()
public MTBLengthEnergyFW2DNonPDE(double mu)
mu - public double deltaE(int x,
int y,
int z,
short newPhase,
MTBLevelsetMembership phi)
MTBGenericEnergyNonPDEphi supplied
from current phase to newPhase.deltaE in class MTBLengthEnergyFWNonPDEnewPhase - new phase to change the pixel/voxel tophi - level set function assumed to be associated with this energy objectprotected abstract double gradMagHeaviside(double a,
double b,
double c)
a - heaviside functtion at (x,y,z)b - heaviside functtion at (x+1,y,z)c - heaviside functtion at (x,y+1,z)public double E(MTBLevelsetMembership phi)
MTBGenericEnergyNonPDEphi.
E in class MTBLengthEnergyFWNonPDEphi - level set function assumed to be associated with this energy objectpublic double E(int x,
int y,
MTBLevelsetMembership phi)
phi. 2D onlyx - y - phi - Copyright © 2010–2020 Martin Luther University Halle-Wittenberg, Institute of Computer Science, Pattern Recognition and Bioinformatics. All rights reserved.