boofcv.alg.flow

Class HornSchunck<T extends boofcv.struct.image.ImageBase,D extends boofcv.struct.image.ImageBase>

java.lang.Object

boofcv.alg.flow.HornSchunck<T,D>

Direct Known Subclasses:

HornSchunck_F32, HornSchunck_U8

public abstract class HornSchunck<T extends boofcv.struct.image.ImageBase,D extends boofcv.struct.image.ImageBase>
extends java.lang.Object

This is Horn-Schunck's well known work [1] for dense optical flow estimation. It is based off the following equation Ex*u + Ey*v + Et = 0, where (u,v) is the estimated flow for a single pixel, and (Ex,Ey) is the pixel's gradient and Et is the grave in intensity value. It is assumed that each pixel maintains a constant intensity and that changes in flow are smooth. This implementation is faithful to the original work and does not make any effort to improve its performance using more modern techniques.

[1] Horn, Berthold K., and Brian G. Schunck. "Determining optical flow." 1981 Technical Symposium East. International Society for Optics and Photonics, 1981.

Field Summary

Fields

Modifier and Type	Field and Description
protected float	alpha2
protected ImageFlow	averageFlow
protected D	derivT
protected D	derivX
protected D	derivY
protected int	numIterations
protected boolean	resetOutput

Constructor Summary

Constructors

Constructor and Description
HornSchunck(float alpha, int numIterations, boofcv.struct.image.ImageType<D> derivType) Constructor

Method Summary

Modifier and Type	Method and Description
protected static void	borderAverageFlow(ImageFlow flow, ImageFlow averageFlow) Computes average flow using an 8-connect neighborhood for the image border
protected static void	computeBorder(ImageFlow flow, ImageFlow averageFlow, int x, int y)
protected abstract void	computeDerivT(T image1, T image2, D derivT)
protected abstract void	computeDerivX(T image1, T image2, D derivX)
protected abstract void	computeDerivY(T image1, T image2, D derivY)
protected abstract void	findFlow(D derivX, D derivY, D derivT, ImageFlow output) Inner function for computing optical flow
protected static ImageFlow.D	getExtend(ImageFlow flow, int x, int y)
protected static void	innerAverageFlow(ImageFlow flow, ImageFlow averageFlow) Computes average flow using an 8-connect neighborhood for the inner image
void	process(T image1, T image2, ImageFlow output) Computes dense optical flow from the first image's gradient and the difference between the second and the first image.
void	setNumIterations(int numIterations) changes the maximum number of iterations

Methods inherited from class java.lang.Object

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

Field Detail

alpha2

protected float alpha2

numIterations

protected int numIterations

averageFlow

protected ImageFlow averageFlow

resetOutput

protected boolean resetOutput

derivX

protected D extends boofcv.struct.image.ImageBase derivX

derivY

protected D extends boofcv.struct.image.ImageBase derivY

derivT

protected D extends boofcv.struct.image.ImageBase derivT

Constructor Detail

HornSchunck

public HornSchunck(float alpha,
                   int numIterations,
                   boofcv.struct.image.ImageType<D> derivType)

Constructor

Parameters:

alpha - Larger values place more importance on flow smoothness consistency over brightness consistency. Try 20

numIterations - Number of iterations. Try 1000

Method Detail

setNumIterations

public void setNumIterations(int numIterations)

changes the maximum number of iterations

Parameters:

numIterations - maximum number of iterations

process

public void process(T image1,
                    T image2,
                    ImageFlow output)

Computes dense optical flow from the first image's gradient and the difference between the second and the first image.

Parameters:

image1 - First image

image2 - Second image

output - Found dense optical flow

computeDerivX

protected abstract void computeDerivX(T image1,
                                      T image2,
                                      D derivX)

computeDerivY

protected abstract void computeDerivY(T image1,
                                      T image2,
                                      D derivY)

computeDerivT

protected abstract void computeDerivT(T image1,
                                      T image2,
                                      D derivT)

findFlow

protected abstract void findFlow(D derivX,
                                 D derivY,
                                 D derivT,
                                 ImageFlow output)

Inner function for computing optical flow

innerAverageFlow

protected static void innerAverageFlow(ImageFlow flow,
                                       ImageFlow averageFlow)

Computes average flow using an 8-connect neighborhood for the inner image

borderAverageFlow

protected static void borderAverageFlow(ImageFlow flow,
                                        ImageFlow averageFlow)

Computes average flow using an 8-connect neighborhood for the image border

computeBorder

protected static void computeBorder(ImageFlow flow,
                                    ImageFlow averageFlow,
                                    int x,
                                    int y)

getExtend

protected static ImageFlow.D getExtend(ImageFlow flow,
                                       int x,
                                       int y)