inra.ijpb.watershed

Class WatershedTransform2D

java.lang.Object

inra.ijpb.watershed.WatershedTransform2D

Direct Known Subclasses:

MarkerControlledWatershedTransform2D

public class WatershedTransform2D
extends Object

Class to apply the watershed algorithm to a 2D image. It allows specifying a binary mask to restrict the regions of interest.

Author:

Ignacio Arganda-Carreras

Field Summary

Fields

Modifier and Type	Field and Description
protected boolean	verbose flag to output the transform steps and their execution times in the log window

Constructor Summary

Constructors

Constructor and Description
WatershedTransform2D(ij.process.ImageProcessor input, ij.process.ImageProcessor mask) Construct a watershed transform
WatershedTransform2D(ij.process.ImageProcessor input, ij.process.ImageProcessor mask, int connectivity) Construct a watershed transform

Method Summary

Modifier and Type	Method and Description
ij.process.ImageProcessor	apply() Apply fast watersheds using flooding simulations, as described by Soille, Pierre, and Luc M.
ij.process.ImageProcessor	apply(double hMin, double hMax) Apply fast watersheds using flooding simulations, as described by Soille, Pierre, and Luc M.
ArrayList<PixelRecord>	extractPixelValues(ij.process.ImageProcessor inputImage, double hMin, double hMax) Extract pixel values from input image such that they have value h, hMin <= h <= hMax.
ij.ImagePlus	getAnimation(double hMin, double hMax) Apply fast watersheds using flooding simulations, as described by Soille, Pierre, and Luc M.
int	getConnectivity() Get the pixel connectivity (4 or 8)
void	setConnectivity(int conn) Set the pixel connectivity (4 or 8)
void	setVerbose(boolean verbose) Set verbose flag

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

verbose

protected boolean verbose

flag to output the transform steps and their execution times in the log window

Constructor Detail

WatershedTransform2D

public WatershedTransform2D(ij.process.ImageProcessor input,
                            ij.process.ImageProcessor mask)

Construct a watershed transform

Parameters:

input - input image (usually a gradient image)

mask - binary mask to restrict the region of interest (null to use whole input image)

WatershedTransform2D

public WatershedTransform2D(ij.process.ImageProcessor input,
                            ij.process.ImageProcessor mask,
                            int connectivity)

Construct a watershed transform

Parameters:

input - input image (usually a gradient image)

mask - binary mask to restrict the region of interest (null to use whole input image)

connectivity - pixel connectivity (4 or 8)

Method Detail

getConnectivity

public int getConnectivity()

Get the pixel connectivity (4 or 8)

Returns:

pixel connectivity

setConnectivity

public void setConnectivity(int conn)

Set the pixel connectivity (4 or 8)

Parameters:

conn - pixel connectivity

setVerbose

public void setVerbose(boolean verbose)

Set verbose flag

Parameters:

verbose - new verbose flag

apply

public ij.process.ImageProcessor apply()

Apply fast watersheds using flooding simulations, as described by Soille, Pierre, and Luc M. Vincent. "Determining watersheds in digital pictures via flooding simulations." Lausanne-DL tentative. International Society for Optics and Photonics, 1990.

Returns:

image of labeled catchment basins (with dams)

apply

public ij.process.ImageProcessor apply(double hMin,
                                       double hMax)

Apply fast watersheds using flooding simulations, as described by Soille, Pierre, and Luc M. Vincent. "Determining watersheds in digital pictures via flooding simulations." Lausanne-DL tentative. International Society for Optics and Photonics, 1990. NOTE: this algorithm may have plateaus in the dams.

Parameters:

hMin - the minimum value for dynamic

hMax - the maximum value for dynamic

Returns:

image of labeled catchment basins (with dams)

extractPixelValues

public ArrayList<PixelRecord> extractPixelValues(ij.process.ImageProcessor inputImage,
                                                 double hMin,
                                                 double hMax)

Extract pixel values from input image such that they have value h, hMin <= h <= hMax. A binary mask is used if it exists.

Parameters:

inputImage - input image

hMin - minimum grayscale height value

hMax - maximum grayscale height value

Returns:

list of input pixel values

getAnimation

public ij.ImagePlus getAnimation(double hMin,
                                 double hMax)

Apply fast watersheds using flooding simulations, as described by Soille, Pierre, and Luc M. Vincent. "Determining watersheds in digital pictures via flooding simulations." Lausanne-DL tentative. International Society for Optics and Photonics, 1990. This implementation uses a binary mask to restrict regions of application if the mask exists. NOTE: this algorithm may have plateaus in the dams.

Parameters:

hMin - minimum grayscale level height

hMax - maximum grayscale level height

Returns:

32-bit stack with the progression of labeled catchment basins (with dams)