lti::opticalFlowHS::parameters Class Reference

the parameters for the class opticalFlowHS More...

#include <ltiOpticalFlowHS.h>

Inheritance diagram for lti::opticalFlowHS::parameters:

[legend]

Collaboration diagram for lti::opticalFlowHS::parameters:

[legend]

List of all members.

Public Member Functions
	parameters ()
	parameters (const parameters &other)
	~parameters ()
const char *	getTypeName () const
parameters &	copy (const parameters &other)
virtual functor::parameters *	clone () const
Public Attributes
int	kernelSize
int	iterations
float	lambda
bool	multiScale
bool	initialize

---

Detailed Description

the parameters for the class opticalFlowHS

---

Constructor & Destructor Documentation

lti::opticalFlowHS::parameters::parameters

(

)

default constructor

Reimplemented from lti::transform::parameters.

lti::opticalFlowHS::parameters::parameters

(

const parameters &

other

)

copy constructor

Parameters:

other

the parameters object to be copied

Reimplemented from lti::transform::parameters.

lti::opticalFlowHS::parameters::~parameters

(

)

[virtual]

destructor

Reimplemented from lti::functor::parameters.

---

Member Function Documentation

virtual functor::parameters* lti::opticalFlowHS::parameters::clone

(

)

const [virtual]

returns a pointer to a clone of the parameters

Reimplemented from lti::transform::parameters.

parameters& lti::opticalFlowHS::parameters::copy

(

const parameters &

other

)

copy the contents of a parameters object

Parameters:

other

the parameters object to be copied

Returns:

a reference to this parameters object

Reimplemented from lti::transform::parameters.

const char* lti::opticalFlowHS::parameters::getTypeName

(

)

const [virtual]

returns name of this type

Reimplemented from lti::transform::parameters.

---

Member Data Documentation

bool lti::opticalFlowHS::parameters::initialize

If you want to hand over previously calculated flow results (u,v) as initial values of iterative flow computation, then this must be set to 'true'.

Otherwise initial values will always be 0. (default = false) Remark: previous results must be passed as function parameters with the 'on copy apply'. (u,v) must be equal sized to input channels ch_t0, ch_t1, or else they will be ignored.

int lti::opticalFlowHS::parameters::iterations

Number of iterations.

iterations = 1,2,... (default=100)

int lti::opticalFlowHS::parameters::kernelSize

Size of kernel for spatial gradient computation.

kernelSize = 3,4,5 (default=3)

float lti::opticalFlowHS::parameters::lambda

Smoothness constraint factor.

lambda > 0.0 (default = 10.0)

The solutions (u,v) of the optical flow equation are ambiguous. The additional "smoothness-constraint" allows to select a solution conditional on the desired smoothness of the flowfield. Smoothness is controlled by parameter 'lambda'. Higher values produce smoother flowfields but also higher error e.

bool lti::opticalFlowHS::parameters::multiScale

Toggles multi scale approach for computation of optical flow.

(default = true)

If multiScale==true, then the two input images are sampled down to several scale levels, before the optical flow is computed iteratively from coarse to fine resolution. This approach allows faster computation and can cope with large motion. Remark: The number of scale levels can vary, since it depends on the number of iterations and the original image size.

---

The documentation for this class was generated from the following file:

• ltiOpticalFlowHS.h