boofcv.alg.filter.derivative

Class GImageDerivativeOps

java.lang.Object

boofcv.alg.filter.derivative.GImageDerivativeOps

public class GImageDerivativeOps
extends java.lang.Object

Generalized operations related to compute different image derivatives.

Constructor Summary

Constructors

Constructor and Description
GImageDerivativeOps()

Method Summary

Modifier and Type	Method and Description
static <I extends ImageSingleBand,D extends ImageSingleBand> AnyImageDerivative<I,D>	createAnyDerivatives(DerivativeType type, java.lang.Class<I> inputType, java.lang.Class<D> derivType) Convenience function for creating an instance of AnyImageDerivative.
static <I extends ImageSingleBand,D extends ImageSingleBand> AnyImageDerivative<I,D>	derivativeForScaleSpace(java.lang.Class<I> inputType, java.lang.Class<D> derivType) Creates an instance of AnyImageDerivative which is intended for use of calculating scale-spaces.
static <I extends ImageSingleBand,D extends ImageSingleBand> java.lang.Class<D>	getDerivativeType(java.lang.Class<I> imageType) Returns the type of image the derivative should be for the specified input type.
static <I extends ImageSingleBand,D extends ImageSingleBand> void	gradient(DerivativeType type, I input, D derivX, D derivY, BorderType borderType) Computes the gradient using the specified image type.
static <D extends ImageSingleBand> void	hessian(DerivativeType type, D derivX, D derivY, D derivXX, D derivYY, D derivXY, BorderType borderType) Computes the hessian from the gradient.
static <I extends ImageSingleBand,D extends ImageSingleBand> void	hessian(DerivativeType type, I input, D derivXX, D derivYY, D derivXY, BorderType borderType) Computes the hessian from the original input image.
static <I extends ImageSingleBand,D extends ImageSingleBand> void	laplace(I input, D output)
static KernelBase	lookupKernelX(DerivativeType type, boolean isInteger) Returns the kernel for finding the X derivative.

Methods inherited from class java.lang.Object

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

Constructor Detail

GImageDerivativeOps

public GImageDerivativeOps()

Method Detail

laplace

public static <I extends ImageSingleBand,D extends ImageSingleBand> void laplace(I input,
                                                                                 D output)

getDerivativeType

public static <I extends ImageSingleBand,D extends ImageSingleBand> java.lang.Class<D> getDerivativeType(java.lang.Class<I> imageType)

Returns the type of image the derivative should be for the specified input type.

Parameters:

imageType - Input image type.

Returns:

Appropriate output image type.

gradient

public static <I extends ImageSingleBand,D extends ImageSingleBand> void gradient(DerivativeType type,
                                                                                  I input,
                                                                                  D derivX,
                                                                                  D derivY,
                                                                                  BorderType borderType)

Computes the gradient using the specified image type.

Type Parameters:

I - Input image type

D - Output image type

Parameters:

type - Type of gradient to compute

input - Input image

derivX - Output. Derivative X

derivY - Output. Derivative Y

borderType - How it should handle borders. null == skip border

hessian

public static <I extends ImageSingleBand,D extends ImageSingleBand> void hessian(DerivativeType type,
                                                                                 I input,
                                                                                 D derivXX,
                                                                                 D derivYY,
                                                                                 D derivXY,
                                                                                 BorderType borderType)

Computes the hessian from the original input image. Only Sobel and Three supported.

Type Parameters:

I - Input image type

D - Output image type

Parameters:

type - Type of gradient to compute

input - Input image

derivXX - Output. Derivative XX

derivYY - Output. Derivative YY

derivXY - Output. Derivative XY

borderType - How it should handle borders. null == skip border

hessian

public static <D extends ImageSingleBand> void hessian(DerivativeType type,
                                                       D derivX,
                                                       D derivY,
                                                       D derivXX,
                                                       D derivYY,
                                                       D derivXY,
                                                       BorderType borderType)

Computes the hessian from the gradient. Only Prewitt, Sobel and Three supported.

Parameters:

type - Type of gradient to compute

derivX - Input derivative X

derivY - Input derivative Y

derivXX - Output. Derivative XX

derivYY - Output. Derivative YY

derivXY - Output. Derivative XY

borderType - How it should handle borders. null == skip border

lookupKernelX

public static KernelBase lookupKernelX(DerivativeType type,
                                       boolean isInteger)

Returns the kernel for finding the X derivative.

Parameters:

type - Type of gradient

isInteger - integer or floating point kernels

Returns:

The kernel. Can be 1D or 2D

createAnyDerivatives

public static <I extends ImageSingleBand,D extends ImageSingleBand> AnyImageDerivative<I,D> createAnyDerivatives(DerivativeType type,
                                                                                                                 java.lang.Class<I> inputType,
                                                                                                                 java.lang.Class<D> derivType)

Convenience function for creating an instance of AnyImageDerivative. This class is an any way to compute any derivative of any order using the specified kernel. It might use more memory or be more expensive the specialized code but is easy to use.

Type Parameters:

I - Image type.

D - Image derivative type.

Parameters:

type - Type of gradient to use

inputType - Type of input image.

derivType - Type of derivative image

Returns:

AnyImageDerivative

derivativeForScaleSpace

public static <I extends ImageSingleBand,D extends ImageSingleBand> AnyImageDerivative<I,D> derivativeForScaleSpace(java.lang.Class<I> inputType,
                                                                                                                    java.lang.Class<D> derivType)

Creates an instance of AnyImageDerivative which is intended for use of calculating scale-spaces. It uses DerivativeType.THREE since it does not blur the image. More typical operators, such as Sobel and Prewitt, blur the image.

Type Parameters:

I - Image type.

D - Image derivative type.

Parameters:

inputType - Type of input image.

derivType - Type of derivative image

Returns:

AnyImageDerivative