features.filterBanks
Class RadialGaussianDerivative
java.lang.Object
features.filterBanks.FilterBank
features.filterBanks.GaussianDerivative
features.filterBanks.RadialGaussianDerivative
public class RadialGaussianDerivative
- extends GaussianDerivative
This class implements a filterbank for radially oriented GaussianDerivative Oriented Filter.
It implements a super-convolver which with one method call
computes multiple convolutions of an image with radial Gaussian derivative kernel at order, scale, and angle.
A radial filter can be 0, 1, 2nd order, and is calculated radially only.
Copyright (c) 1999-2005, Michael Abramoff. All rights reserved.
|
Method Summary |
Feature |
filter(int filternumber,
int[] image,
int width,
java.lang.String extraText)
Apply a filter to the image. |
int |
getNumber()
Return the number of filters in this filter bank. |
java.lang.String |
name(int filternumber,
java.lang.String preString)
Return a descriptive String for this filternumber. |
float[] |
R(float[] image,
int width,
int n,
double scale,
boolean radial)
Compute a radial L filter of order, scale and orientation. |
| Methods inherited from class java.lang.Object |
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
RadialGaussianDerivative
public RadialGaussianDerivative(float[] scales,
int nrOrders,
int[] channels)
getNumber
public int getNumber()
- Return the number of filters in this filter bank.
- Overrides:
getNumber in class GaussianDerivative
name
public java.lang.String name(int filternumber,
java.lang.String preString)
- Description copied from class:
GaussianDerivative
- Return a descriptive String for this filternumber.
- Overrides:
name in class GaussianDerivative
filter
public Feature filter(int filternumber,
int[] image,
int width,
java.lang.String extraText)
- Apply a filter to the image.
The filter is determined by the filternumber which indicates which scale, order and orientation to use.
filternumber is a consecutive number that identifies a filter with a specific scale, derivative order and orientation.
- Overrides:
filter in class GaussianDerivative
R
public float[] R(float[] image,
int width,
int n,
double scale,
boolean radial)
- Compute a radial L filter of order, scale and orientation.
- Parameters:
image - the imagewidth - the width of image in pixels.n - the order of the derivativesscale - the scale of the GaussianDerivative used to compute the derivatives.radial - the direction of the derivative, either radial or tangential.
- Returns:
- a float[][] with the resulting convolution.