public class RandomWalkOptions extends Object
Modifier and Type | Field and Description |
---|---|
double |
alpha
Alpha penalises pixels whose intensities are far away from the mean seed intensity.
|
double |
beta
Beta penalises pixels located at an edge, i.e.
|
double |
dt
Timestep, if dt=1 we are at the limit of CFL stability.
|
BinaryFilters |
finalFilter
Reference to filter used on final processing.
|
double[] |
gamma
Gamma is the strength of competition between foreground and background.
|
BinaryFilters |
intermediateFilter
Reference to filter used for filtering results between sweeps.
|
int |
iter
Maximum number of Euler iterations.
|
int |
localMeanMaskSize
Size of mask for local mean algorithm (odd).
|
boolean |
maskLimit
If true, RW mask will be cut by AC mask.
|
double[] |
relim
Upper relative error limit used as stopping criterion.
|
boolean |
useLocalMean
true if local mean algorithm is used.
|
public double alpha
public double beta
public double[] gamma
gamma[1]==0 disables second sweep.
public int iter
public double dt
public double[] relim
Contains errors rot two steps.
public boolean useLocalMean
If localMean is used, the seeds provided to
RandomWalkSegmentation.run(Seeds)
must have RandomWalkSegmentation.SeedTypes.ROUGHMASK
entry.
public int localMeanMaskSize
public transient BinaryFilters intermediateFilter
null value switches off second sweep as well as gamma[1]==0. To switch off filtering between
sweeps use
BinaryFilters.EmptyMorpho
public transient BinaryFilters finalFilter
null value switches off final filtering.
public boolean maskLimit
public RandomWalkOptions()
public RandomWalkOptions(Double alpha, Double beta, Double gamma1, Double gamma2, Integer iter, Double dt, Double[] relim, Boolean useLocalMean, Integer localMeanMaskSize)
alpha
- alphabeta
- betagamma1
- gamma1gamma2
- gamma2iter
- iterdt
- dtrelim
- relim (will be copied)useLocalMean
- useLocalMeanlocalMeanMaskSize
- localMeanMaskSizeCopyright © 2002–2019 Department of Computer Science, Warwick University. All rights reserved.