Package inra.ijpb.binary

Class BinaryImages

java.lang.Object

inra.ijpb.binary.BinaryImages

public class BinaryImages
extends java.lang.Object

A collection of static methods for operating on binary images (2D/3D). Some of the methods need the LabelImages class.

Author:

David Legland

See Also:

LabelImages

Method Summary

Modifier and Type	Method	Description
static ij.process.ImageProcessor	areaOpening(ij.process.ImageProcessor image, int nPixelMin)	Applies area opening on a binary image: creates a new binary image that contains only particles with at least the specified number of pixels.
static ij.ImagePlus	binarize(ij.ImagePlus imagePlus)	Converts a grayscale 2D or 3D image into a binary image by setting non-zero elements to 255.
static ij.ImageStack	binarize(ij.ImageStack image)	Converts a grayscale 3D image into a binary 3D image by setting non-zero voxels to 255.
static ij.process.ImageProcessor	binarize(ij.process.ImageProcessor image)	Converts a grayscale 2D image into a binary 2D image by setting non-zero pixels to 255.
static ij.ImagePlus	binarizeBackground(ij.ImagePlus imagePlus)	Converts a grayscale 2D or 3D image into a binary image by setting zero elements to 255, and non zero ones to 0.
static ij.ImageStack	binarizeBackground(ij.ImageStack image)	Converts a grayscale 3D image into a binary 3D image by setting zero elements to 255, and non zero ones to 0.
static ij.process.ImageProcessor	binarizeBackground(ij.process.ImageProcessor image)	Converts a grayscale 2D image into a binary 2D image by setting zero elements to 255, and non zero ones to 0.
static ij.ImagePlus	componentsLabeling(ij.ImagePlus imagePlus, int conn, int bitDepth)	Computes the labels in the binary 2D or 3D image contained in the given ImagePlus, and computes the maximum label to set up the display range of the resulting ImagePlus.
static ij.ImageStack	componentsLabeling(ij.ImageStack image, int conn, int bitDepth)	Computes the labels of the connected components in the given 3D binary image.
static ij.process.ImageProcessor	componentsLabeling(ij.process.ImageProcessor image, int conn, int bitDepth)	Computes the labels of the connected components in the given planar binary image.
static int	countForegroundPixels(ij.process.ImageProcessor image)	Counts the number of foreground pixels.
static int	countForegroundVoxels(ij.ImageStack image)	Counts the number of foreground voxels.
static ij.ImagePlus	distanceMap(ij.ImagePlus imagePlus)	Computes the distance map (or distance transform) from a binary image processor.
static ij.ImageStack	distanceMap(ij.ImageStack image)	Computes the distance map from a binary 3D image.
static ij.ImageStack	distanceMap(ij.ImageStack image, float[] weights, boolean normalize)	Computes the distance map from a binary 3D image.
static ij.ImageStack	distanceMap(ij.ImageStack image, short[] weights, boolean normalize)	Computes the distance map from a binary 3D image.
static ij.ImageStack	distanceMap(ij.ImageStack image, ChamferMask3D chamferMask, boolean floatingPoint, boolean normalize)	Computes the distance map from a binary 3D image.
static ij.process.ImageProcessor	distanceMap(ij.process.ImageProcessor image)	Computes the distance map (or distance transform) from a binary image processor.
static ij.process.FloatProcessor	distanceMap(ij.process.ImageProcessor image, float[] weights, boolean normalize)	Computes the distance map from a binary image processor, by specifying weights and normalization.
static ij.process.ShortProcessor	distanceMap(ij.process.ImageProcessor image, short[] weights, boolean normalize)	Computes the distance map from a binary image processor, by specifying weights and normalization.
static ij.process.ImageProcessor	distanceMap(ij.process.ImageProcessor image, ChamferMask2D mask, boolean floatingPoint, boolean normalize)	Computes the distance map from a binary image processor, by specifying weights and normalization.
static ij.ImagePlus	geodesicDistanceMap(ij.ImagePlus markerPlus, ij.ImagePlus maskPlus)	Computes the geodesic distance transform (or geodesic distance map) of a binary image of marker, constrained to a binary mask.
static ij.process.ImageProcessor	geodesicDistanceMap(ij.process.ImageProcessor marker, ij.process.ImageProcessor mask)	Computes the geodesic distance transform (or geodesic distance map) of a binary image of marker, constrained to a binary mask.
static ij.process.ImageProcessor	geodesicDistanceMap(ij.process.ImageProcessor marker, ij.process.ImageProcessor mask, float[] weights, boolean normalize)	Computes the geodesic distance transform (or geodesic distance map) of a binary image of marker, constrained to a binary mask.
static ij.process.ImageProcessor	geodesicDistanceMap(ij.process.ImageProcessor marker, ij.process.ImageProcessor mask, short[] weights, boolean normalize)	Computes the geodesic distance transform (or geodesic distance map) of a binary image of marker, constrained to a binary mask.
static boolean	isBinaryImage(ij.ImagePlus image)	Check if input image is binary (8-bit with only 0 or 255 values)
static boolean	isBinaryImage(ij.process.ImageProcessor image)	Check if input image is binary (8-bit with only 0 or 255 values)
static ij.ImagePlus	keepLargestRegion(ij.ImagePlus imagePlus)	Returns a binary image that contains only the largest region.
static ij.ImageStack	keepLargestRegion(ij.ImageStack image)	Returns a binary image that contains only the largest label.
static ij.process.ImageProcessor	keepLargestRegion(ij.process.ImageProcessor image)	Returns a binary image that contains only the largest region.
static void	removeLargestRegion(ij.ImagePlus imagePlus)	Returns a binary image in which the largest region has been replaced by background.
static ij.ImageStack	removeLargestRegion(ij.ImageStack image)	Returns a binary image in which the largest region has been replaced by the background value.
static ij.process.ImageProcessor	removeLargestRegion(ij.process.ImageProcessor image)	Returns a binary image in which the largest region has been replaced by the background value.
static ij.ImagePlus	sizeOpening(ij.ImagePlus imagePlus, int minElementCount)	Applies size opening on a binary 2D or 3D image.
static ij.process.ImageProcessor	skeleton(ij.process.ImageProcessor image)	Computes the skeleton of a binary image, and returns another binary image.
static ij.ImageStack	volumeOpening(ij.ImageStack image, int nVoxelMin)	Applies area opening on a binary image: creates a new binary image that contains only particle with at least the specified number of voxels.

Methods inherited from class java.lang.Object

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

Method Details

isBinaryImage

public static final boolean isBinaryImage(ij.ImagePlus image)

Check if input image is binary (8-bit with only 0 or 255 values)

Parameters:

image - input image

Returns:

true if image is binary

isBinaryImage

public static final boolean isBinaryImage(ij.process.ImageProcessor image)

Check if input image is binary (8-bit with only 0 or 255 values)

Parameters:

image - input image

Returns:

true if image is binary

countForegroundPixels

public static final int countForegroundPixels(ij.process.ImageProcessor image)

Counts the number of foreground pixels. The foreground pixels are the pixels with a value greater than zero.

Parameters:

image - a binary image with zero values for background

Returns:

the number of pixels with value greater than zero

countForegroundVoxels

public static final int countForegroundVoxels(ij.ImageStack image)

Counts the number of foreground voxels. The foreground voxels are the voxels with a value greater than zero.

Parameters:

image - a binary 3D image with zero values for background

Returns:

the number of voxels with value greater than zero

componentsLabeling

public static final ij.ImagePlus componentsLabeling(ij.ImagePlus imagePlus,
int conn,
int bitDepth)

Computes the labels in the binary 2D or 3D image contained in the given ImagePlus, and computes the maximum label to set up the display range of the resulting ImagePlus.

Parameters:

imagePlus - contains the 3D binary image stack

conn - the connectivity, either 4 or 8 for planar images, or 6 or 26 for 3D images

bitDepth - the number of bits used to create the result image (8, 16 or 32)

Returns:

an ImagePlus containing the label of each connected component.

Throws:

java.lang.RuntimeException - if the number of labels reaches the maximum number that can be represented with this bitDepth

See Also:

ConnectedComponentsLabeling, ConnectedComponentsLabeling3D, FloodFill

componentsLabeling

public static final ij.process.ImageProcessor componentsLabeling(ij.process.ImageProcessor image,
int conn,
int bitDepth)

Computes the labels of the connected components in the given planar binary image. The type of result is controlled by the bitDepth option. Uses a Flood-fill type algorithm.

Parameters:

image - contains the binary image (any type is accepted)

conn - the connectivity, either 4 or 8

bitDepth - the number of bits used to create the result image (8, 16 or 32)

Returns:

a new instance of ImageProcessor containing the label of each connected component.

Throws:

java.lang.RuntimeException - if the number of labels reaches the maximum number that can be represented with this bitDepth

See Also:

ConnectedComponentsLabeling

componentsLabeling

public static final ij.ImageStack componentsLabeling(ij.ImageStack image,
int conn,
int bitDepth)

Computes the labels of the connected components in the given 3D binary image. The type of result is controlled by the bitDepth option. Uses a Flood-fill type algorithm.

Parameters:

image - contains the 3D binary image (any type is accepted)

conn - the connectivity, either 6 or 26

bitDepth - the number of bits used to create the result stack (8, 16 or 32)

Returns:

a new instance of ImageStack containing the label of each connected component.

Throws:

java.lang.RuntimeException - if the number of labels reaches the maximum number that can be represented with this bitDepth

See Also:

ConnectedComponentsLabeling3D

distanceMap

public static final ij.ImagePlus distanceMap(ij.ImagePlus imagePlus)

Computes the distance map (or distance transform) from a binary image processor. Distance is computed for each foreground (white) pixel or voxel, as the chamfer distance to the nearest background (black) pixel or voxel.

Parameters:

imagePlus - an ImagePlus object containing a binary or label image

Returns:

a new ImagePlus containing the distance map

See Also:

DistanceTransform, DistanceTransform3D

distanceMap

public static final ij.process.ImageProcessor distanceMap(ij.process.ImageProcessor image)

Computes the distance map (or distance transform) from a binary image processor. Distance is computed for each foreground (white) pixel, as the chamfer distance to the nearest background (black) pixel.

This method uses default 5x5 weights, and normalizes the resulting map. Result is given in a new instance of ShortProcessor.

Parameters:

image - the input binary or label image

Returns:

a new ImageProcessor containing the distance map result

distanceMap

public static final ij.process.ImageProcessor distanceMap(ij.process.ImageProcessor image,
ChamferMask2D mask,
boolean floatingPoint,
boolean normalize)

Computes the distance map from a binary image processor, by specifying weights and normalization.

Distance is computed for each foreground (white) pixel, as the chamfer distance to the nearest background (black) pixel. Result is given in a new instance of FloatProcessor.

Parameters:

image - the input binary or label image

mask - the chamfer mask used to propagate distances

floatingPoint - indicates if the computation should be performed using floating point computation

normalize - indicates whether the resulting distance map should be normalized (divide distances by the first chamfer weight)

Returns:

the distance map obtained after applying the distance transform

distanceMap

public static final ij.process.ShortProcessor distanceMap(ij.process.ImageProcessor image,
short[] weights,
boolean normalize)

Computes the distance map from a binary image processor, by specifying weights and normalization.

Distance is computed for each foreground (white) pixel, as the chamfer distance to the nearest background (black) pixel. Result is given as a new instance of ShortProcessor.

Parameters:

image - the input binary or label image

weights - an array of chamfer weights, with at least two values

normalize - indicates whether the resulting distance map should be normalized (divide distances by the first chamfer weight)

Returns:

the distance map obtained after applying the distance transform

distanceMap

public static final ij.process.FloatProcessor distanceMap(ij.process.ImageProcessor image,
float[] weights,
boolean normalize)

Computes the distance map from a binary image processor, by specifying weights and normalization.

Distance is computed for each foreground (white) pixel, as the chamfer distance to the nearest background (black) pixel. Result is given in a new instance of FloatProcessor.

Parameters:

image - the input binary or label image

weights - an array of chamfer weights, with at least two values

normalize - indicates whether the resulting distance map should be normalized (divide distances by the first chamfer weight)

Returns:

the distance map obtained after applying the distance transform

distanceMap

public static final ij.ImageStack distanceMap(ij.ImageStack image)

Computes the distance map from a binary 3D image. Distance is computed for each foreground (white) pixel, as the chamfer distance to the nearest background (black) pixel.

Parameters:

image - the input binary image

Returns:

the distance map obtained after applying the distance transform

distanceMap

public static final ij.ImageStack distanceMap(ij.ImageStack image,
ChamferMask3D chamferMask,
boolean floatingPoint,
boolean normalize)

Computes the distance map from a binary 3D image. Distance is computed for each foreground (white) pixel, as the chamfer distance to the nearest background (black) pixel.

Parameters:

image - the input binary image

chamferMask - the chamfer mask used for propagating distances

floatingPoint - indicates if the distane propagation should use floating point computation

normalize - indicates whether the resulting distance map should be normalized (divide distances by the first chamfer weight)

Returns:

the distance map obtained after applying the distance transform

distanceMap

public static final ij.ImageStack distanceMap(ij.ImageStack image,
short[] weights,
boolean normalize)

Computes the distance map from a binary 3D image. Distance is computed for each foreground (white) pixel, as the chamfer distance to the nearest background (black) pixel.

Parameters:

image - the input binary image

weights - an array of chamfer weights, with at least three values

normalize - indicates whether the resulting distance map should be normalized (divide distances by the first chamfer weight)

Returns:

the distance map obtained after applying the distance transform

distanceMap

public static final ij.ImageStack distanceMap(ij.ImageStack image,
float[] weights,
boolean normalize)

Computes the distance map from a binary 3D image. Distance is computed for each foreground (white) pixel, as the chamfer distance to the nearest background (black) pixel.

Parameters:

image - the input 3D binary image

weights - an array of chamfer weights, with at least three values

normalize - indicates whether the resulting distance map should be normalized (divide distances by the first chamfer weight)

Returns:

the distance map obtained after applying the distance transform

geodesicDistanceMap

public static final ij.ImagePlus geodesicDistanceMap(ij.ImagePlus markerPlus,
ij.ImagePlus maskPlus)

Computes the geodesic distance transform (or geodesic distance map) of a binary image of marker, constrained to a binary mask. Returns the result in a new instance of ImagePlus.

Parameters:

markerPlus - the image containing the marker

maskPlus - the image containing the marker

Returns:

the geodesic distance map in a new ImagePlus

geodesicDistanceMap

public static final ij.process.ImageProcessor geodesicDistanceMap(ij.process.ImageProcessor marker,
ij.process.ImageProcessor mask)

Computes the geodesic distance transform (or geodesic distance map) of a binary image of marker, constrained to a binary mask. Returns the result in a new instance of ShortProcessor.

Parameters:

marker - the binary image of marker

mask - the binary image of mask

Returns:

the geodesic distance map in a new ImageProcessor

geodesicDistanceMap

public static final ij.process.ImageProcessor geodesicDistanceMap(ij.process.ImageProcessor marker,
ij.process.ImageProcessor mask,
short[] weights,
boolean normalize)

Computes the geodesic distance transform (or geodesic distance map) of a binary image of marker, constrained to a binary mask. Returns the result in a new instance of ShortProcessor.

Parameters:

marker - the binary image of marker

mask - the binary image of mask

weights - an array of chamfer weights, with at least two values

normalize - indicates whether the resulting distance map should be normalized (divide distances by the first chamfer weight)

Returns:

the geodesic distance map in a new ImageProcessor

geodesicDistanceMap

public static final ij.process.ImageProcessor geodesicDistanceMap(ij.process.ImageProcessor marker,
ij.process.ImageProcessor mask,
float[] weights,
boolean normalize)

Computes the geodesic distance transform (or geodesic distance map) of a binary image of marker, constrained to a binary mask. Returns the result in a new instance of FloatProcessor.

Parameters:

marker - the binary image of marker

mask - the binary image of mask

weights - an array of chamfer weights, with at least two values

normalize - indicates whether the resulting distance map should be normalized (divide distances by the first chamfer weight)

Returns:

the geodesic distance map in a new ImageProcessor

sizeOpening

public static final ij.ImagePlus sizeOpening(ij.ImagePlus imagePlus,
int minElementCount)

Applies size opening on a binary 2D or 3D image. The method creates a new binary image that contains only particles with at least the specified number of pixels.

Parameters:

imagePlus - the binary 2D or 3D image containing individual particles

minElementCount - the minimum number of pixels or voxels required to keep a particle

Returns:

a new binary image containing only the selected particles

See Also:

LabelImages.sizeOpening(ImagePlus, int)

areaOpening

public static final ij.process.ImageProcessor areaOpening(ij.process.ImageProcessor image,
int nPixelMin)

Applies area opening on a binary image: creates a new binary image that contains only particles with at least the specified number of pixels.

Parameters:

image - the binary image containing individual particles

nPixelMin - the minimum number of pixels required to keep a particle

Returns:

a new binary image containing only the selected particles

See Also:

LabelImages.areaOpening(ImageProcessor, int)

volumeOpening

public static final ij.ImageStack volumeOpening(ij.ImageStack image,
int nVoxelMin)

Applies area opening on a binary image: creates a new binary image that contains only particle with at least the specified number of voxels.

Parameters:

image - the 3D binary image containing individual particles

nVoxelMin - the minimum number of voxels required to keep a particle

Returns:

a new binary image containing only the selected particles

See Also:

LabelImages.volumeOpening(ImageStack, int)

keepLargestRegion

public static final ij.ImagePlus keepLargestRegion(ij.ImagePlus imagePlus)

Returns a binary image that contains only the largest region.

Parameters:

imagePlus - an instance of imagePlus that contains a binary image

Returns:

a new ImagePlus containing only the largest region from original image

keepLargestRegion

public static final ij.process.ImageProcessor keepLargestRegion(ij.process.ImageProcessor image)

Returns a binary image that contains only the largest region.

Parameters:

image - a binary image containing several individual regions

Returns:

a new binary image containing only the largest region from original image

keepLargestRegion

public static final ij.ImageStack keepLargestRegion(ij.ImageStack image)

Returns a binary image that contains only the largest label.

Parameters:

image - a binary 3D image containing several individual regions

Returns:

a new binary image containing only the largest region from original image

removeLargestRegion

public static final void removeLargestRegion(ij.ImagePlus imagePlus)

Returns a binary image in which the largest region has been replaced by background. Works for both 2D and 3D images.

Parameters:

imagePlus - a binary image containing several individual particles

removeLargestRegion

public static final ij.process.ImageProcessor removeLargestRegion(ij.process.ImageProcessor image)

Returns a binary image in which the largest region has been replaced by the background value.

Parameters:

image - a binary image containing several individual particles

Returns:

a new binary image containing all the regions from original image but the largest one

removeLargestRegion

public static final ij.ImageStack removeLargestRegion(ij.ImageStack image)

Returns a binary image in which the largest region has been replaced by the background value.

Parameters:

image - a binary 3D image containing several individual particles

Returns:

a new binary image containing all the regions from original image but the largest one

binarize

public static final ij.ImagePlus binarize(ij.ImagePlus imagePlus)

Converts a grayscale 2D or 3D image into a binary image by setting non-zero elements to 255.

Parameters:

imagePlus - an ImagePlus object containing a gray scale image

Returns:

a binary image containing 255 for all non-zero elements of original image

binarize

public static final ij.process.ImageProcessor binarize(ij.process.ImageProcessor image)

Converts a grayscale 2D image into a binary 2D image by setting non-zero pixels to 255.

Parameters:

image - a gray scale image

Returns:

a binary image containing 255 for all non-zero elements of original image

binarize

public static final ij.ImageStack binarize(ij.ImageStack image)

Converts a grayscale 3D image into a binary 3D image by setting non-zero voxels to 255.

Parameters:

image - a gray scale 3D image

Returns:

a binary image containing 255 for all non-zero elements of original image

binarizeBackground

public static final ij.ImagePlus binarizeBackground(ij.ImagePlus imagePlus)

Converts a grayscale 2D or 3D image into a binary image by setting zero elements to 255, and non zero ones to 0.

Parameters:

imagePlus - an ImagePlus object containing a gray scale image

Returns:

a binary image containing 255 for all non-zero elements of original image

binarizeBackground

public static final ij.process.ImageProcessor binarizeBackground(ij.process.ImageProcessor image)

Converts a grayscale 2D image into a binary 2D image by setting zero elements to 255, and non zero ones to 0.

Parameters:

image - a gray scale image

Returns:

a binary image containing 255 for all non-zero elements of original image

binarizeBackground

public static final ij.ImageStack binarizeBackground(ij.ImageStack image)

Converts a grayscale 3D image into a binary 3D image by setting zero elements to 255, and non zero ones to 0.

Parameters:

image - a gray scale 3D image

Returns:

a binary image containing 255 for all non-zero elements of original image

skeleton

public static final ij.process.ImageProcessor skeleton(ij.process.ImageProcessor image)

Computes the skeleton of a binary image, and returns another binary image. Uses an adaptation of the algorithm from ImageJ.

Parameters:

image - a binary image

Returns:

a binary image containing 255 for skeleton elements