imagingbook.lib.image

Class IntegralImage

java.lang.Object

imagingbook.lib.image.IntegralImage

public class IntegralImage
extends Object

This class represents an 'integral image' or 'summed area table' [Crow, 1984], as described in the book (see 2nd English ed. 2016, Sec. 3.8). Currently only implemented for images of type ByteProcessor.

Version:

2015/11/15

Constructor Summary

Constructors

Constructor and Description
IntegralImage(ij.process.ByteProcessor I) Creates a new integral image from pixel values in a ByteProcessor.
IntegralImage(int[][] I) Creates a new integral image from pixel values in a 2D int-array.

Method Summary

Modifier and Type	Method and Description
long	getBlockSum1(int ua, int va, int ub, int vb) Calculates the sum of the pixel values in the rectangle R, specified by the corner points a = (ua, va) and b = (b1, vb).
long	getBlockSum2(int ua, int va, int ub, int vb) Calculates the sum of the squared pixel values in the rectangle R, specified by the corner points a = (ua, va) and b = (b1, vb).
double	getMean(int ua, int va, int ub, int vb) Calculates the mean of the image values in the specified rectangle.
long[][]	getS1() Returns the summed area table of pixel values (Sigma_1).
long[][]	getS2() Returns the summed area table of squared pixel values (Sigma_2).
int	getSize(int u0, int v0, int u1, int v1)
double	getVariance(int ua, int va, int ub, int vb) Calculates the variance of the image values in the specified rectangle.

Methods inherited from class java.lang.Object

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

Constructor Detail

IntegralImage

public IntegralImage(int[][] I)

Creates a new integral image from pixel values in a 2D int-array.

Parameters:

I - pixel values

IntegralImage

public IntegralImage(ij.process.ByteProcessor I)

Creates a new integral image from pixel values in a ByteProcessor.

Parameters:

I - input image

Method Detail

getS1

public long[][] getS1()

Returns the summed area table of pixel values (Sigma_1).

Returns:

Array of Sigma_1 values

getS2

public long[][] getS2()

Returns the summed area table of squared pixel values (Sigma_2).

Returns:

Array of Sigma_2 values

getBlockSum1

public long getBlockSum1(int ua,
                         int va,
                         int ub,
                         int vb)

Calculates the sum of the pixel values in the rectangle R, specified by the corner points a = (ua, va) and b = (b1, vb).

Parameters:

ua - leftmost position in R

va - top position in R

ub - rightmost position in R

vb - bottom position in R

Returns:

the first-order block sum (S1(R)) inside the specified rectangle or zero if the rectangle is empty.

getBlockSum2

public long getBlockSum2(int ua,
                         int va,
                         int ub,
                         int vb)

Calculates the sum of the squared pixel values in the rectangle R, specified by the corner points a = (ua, va) and b = (b1, vb).

Parameters:

ua - leftmost position in R

va - top position in R

ub - rightmost position in R

vb - bottom position in R

Returns:

the second-order block sum (S2(R)) inside the specified rectangle or zero if the rectangle is empty.

getSize

public int getSize(int u0,
                   int v0,
                   int u1,
                   int v1)

getMean

public double getMean(int ua,
                      int va,
                      int ub,
                      int vb)

Calculates the mean of the image values in the specified rectangle.

Parameters:

ua - leftmost position in R

va - top position in R

ub - rightmost position in R (u1 >= u0)

vb - bottom position in R (v1 >= v0)

Returns:

the mean value for the specified rectangle

getVariance

public double getVariance(int ua,
                          int va,
                          int ub,
                          int vb)

Calculates the variance of the image values in the specified rectangle. parameters.

Parameters:

ua - leftmost position in R

va - top position in R

ub - rightmost position in R (u1 >= u0)

vb - bottom position in R (v1 >= v0)

Returns:

the variance for the specified rectangle