volume
Class Gaussian

java.lang.Object
  |
  +--volume.Kernel
        |
        +--volume.Kernel1D
              |
              +--volume.Gaussian

Direct Known Subclasses:

DoG

public class Gaussian

extends Kernel1D

This is a 1D separated 0-th order Gaussian convolution kernel. G(x) = 1.0 / (Math.sqrt(2 Math.PI) * sigma)) * e^(-l * x / (2.0 * sigma^2)) (c) 1999-2002 Michael Abramoff. All rights reserved.

Field Summary

protected double

sigma

Fields inherited from class volume.Kernel1D

k

Fields inherited from class volume.Kernel

halfwidth

Constructor Summary

Gaussian()

Gaussian(double sigma)
Create a Gaussian kernel from standard deviation sigma.

Gaussian(int width)
Deprecated.

Method Summary

protected double	function(double x) Compute Difference of Gaussian (DoG) function at x symmetric around 0.
double	getSigma()
java.lang.String	toString()

Methods inherited from class volume.Kernel1D

kernelToString

Methods inherited from class volume.Kernel

support

Methods inherited from class java.lang.Object

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

Field Detail

sigma

protected double sigma

Constructor Detail

Gaussian

public Gaussian()

Gaussian

public Gaussian(double sigma)

Create a Gaussian kernel from standard deviation sigma.

Parameters:

sigma, - the standard deviation of the associated probability function of the Gaussian function.

Gaussian

public Gaussian(int width)

Deprecated.

Preset Gaussian for a specific kernel width. 3 and 5 widths only!

Method Detail

getSigma

public double getSigma()

function

protected double function(double x)

Compute Difference of Gaussian (DoG) function at x symmetric around 0.

Parameters:

x - the x position

Returns:

a double with the value of the Gaussian at x.

toString

public java.lang.String toString()

Overrides:

toString in class Kernel