org.openimaj.image.analysis.algorithm

Class EuclideanDistanceTransform

java.lang.Object

org.openimaj.image.analysis.algorithm.EuclideanDistanceTransform

All Implemented Interfaces:

ImageAnalyser<FImage>

public class EuclideanDistanceTransform
extends Object
implements ImageAnalyser<FImage>

See http://people.cs.uchicago.edu/~pff/papers/dt.pdf An efficient euclidean distance transform applicable to all greyscale images. The distance of each pixel to the closest valid pixel is given. In this case a pixel is considered valid when it is less than Float.MAX_VALUE.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Constructor Summary

Constructors

Constructor and Description
EuclideanDistanceTransform()

Method Summary

Modifier and Type	Method and Description
void	analyseImage(FImage image) Analyse an image.
protected static void	DT1D(float[] f, float[] d, int[] v, int[] l, float[] z)
FImage	getDistances()
int[][]	getIndices()
static void	main(String[] args) Test the distance transform
static void	squaredEuclideanDistance(FImage image, FImage distances, int[][] indices) The static function which underlies EuclideanDistanceTransform.
static void	squaredEuclideanDistanceBinary(FImage image, FImage distances, int[][] indices) Calculate the squared euclidean distance transform of a binary image with foreground pixels set to 1 and background set to 0.

Methods inherited from class java.lang.Object

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

Constructor Detail

EuclideanDistanceTransform

public EuclideanDistanceTransform()

Method Detail

analyseImage

public void analyseImage(FImage image)

Description copied from interface: ImageAnalyser

Analyse an image.

Specified by:

analyseImage in interface ImageAnalyser<FImage>

Parameters:

image - The image to process in place.

getDistances

public FImage getDistances()

Returns:

the distance transformed image (unnormalised)

getIndices

public int[][] getIndices()

Returns:

the indecies of the closest pixel to any given pixel

DT1D

protected static void DT1D(float[] f,
                           float[] d,
                           int[] v,
                           int[] l,
                           float[] z)

squaredEuclideanDistanceBinary

public static void squaredEuclideanDistanceBinary(FImage image,
                                                  FImage distances,
                                                  int[][] indices)

Calculate the squared euclidean distance transform of a binary image with foreground pixels set to 1 and background set to 0.

Parameters:

image - the image to be transformed.

distances - the distance of each pixel to the closest 1-pixel

indices - the index of the closes valid pixel

squaredEuclideanDistance

public static void squaredEuclideanDistance(FImage image,
                                            FImage distances,
                                            int[][] indices)

The static function which underlies EuclideanDistanceTransform. Provide an image, fill distances and indices with the distance image and the closest pixel indices. Typically, for the binary case, valid pixels are set to 0 and invalid pixels are set to Float.MAX_VALUE or Float.POSITIVE_INFINITY.

Parameters:

image - the image to be transformed. Each pixel is considered valid except those of value Float.MAX_VALUE

distances - the distance of each pixel to the closest non-Float.MAX_VALUE pixel

indices - the index of the closes valid pixel

main

public static void main(String[] args)
                 throws IOException

Test the distance transform

Parameters:

args -

Throws:

IOException