PCL
pcl::FFTTranslation Class Reference

FFT registration engine: translation More...

#include <FFTRegistration.h>

+ Inheritance diagram for pcl::FFTTranslation:

Public Member Functions

 FFTTranslation ()=default
 
 ~FFTTranslation () override
 
bool AreLargeTranslationsEnabled () const
 
const FPointDelta () const
 
float DeltaX () const
 
float DeltaY () const
 
void DisableLargeTranslations (bool disable=true)
 
void EnableLargeTranslations (bool enable=true)
 
float Peak () const
 
- Public Member Functions inherited from pcl::FFTRegistrationEngine
 FFTRegistrationEngine ()=default
 
 FFTRegistrationEngine (const FFTRegistrationEngine &)=default
 
 FFTRegistrationEngine (FFTRegistrationEngine &&x)
 
virtual ~FFTRegistrationEngine ()
 
const ComplexImageDFTOfReferenceImage () const
 
void Evaluate (const ImageVariant &image)
 
template<class P >
void Evaluate (const pcl::GenericImage< P > &image)
 
void Initialize (const ImageVariant &image)
 
template<class P >
void Initialize (const pcl::GenericImage< P > &image)
 
bool IsInitialized () const
 
void Reset ()
 

Detailed Description

FFTTranslation evaluates translation image registration parameters. It computes the offsets dx and dy required to register a target image on a reference image, using an FFT-based phase correlation algorithm.

See also
FFTRotationAndScaling, FFTRegistrationEngine

Definition at line 312 of file FFTRegistration.h.

Constructor & Destructor Documentation

◆ FFTTranslation()

pcl::FFTTranslation::FFTTranslation ( )
default

Constructs an FFTTranslation object.

◆ ~FFTTranslation()

pcl::FFTTranslation::~FFTTranslation ( )
inlineoverride

Destroys an FFTTranslation object.

Definition at line 324 of file FFTRegistration.h.

Member Function Documentation

◆ AreLargeTranslationsEnabled()

bool pcl::FFTTranslation::AreLargeTranslationsEnabled ( ) const
inline

Returns true iff this registration engine can evaluate translations greater than or equal to one half of the largest dimension of the reference image.

See also
EnableLargeTranslations(), DisableLargeTranslations()

Definition at line 335 of file FFTRegistration.h.

◆ Delta()

const FPoint& pcl::FFTTranslation::Delta ( ) const
inline

Returns the evaluated translation increments in pixels.

The returned Point object contains the required offsets in the X and Y directions to register the last evaluated target image on the reference image set upon initialization.

See also
DeltaX(), DeltaY()

Definition at line 386 of file FFTRegistration.h.

◆ DeltaX()

float pcl::FFTTranslation::DeltaX ( ) const
inline

Returns the evaluated translation increment in the X direction, in pixels.

See also
DeltaY(), Delta()

Definition at line 397 of file FFTRegistration.h.

◆ DeltaY()

float pcl::FFTTranslation::DeltaY ( ) const
inline

Returns the evaluated translation increment in the Y direction, in pixels.

See also
DeltaX(), Delta()

Definition at line 408 of file FFTRegistration.h.

◆ DisableLargeTranslations()

void pcl::FFTTranslation::DisableLargeTranslations ( bool  disable = true)
inline

Disables or enables evaluation of large translations (>= one half of the reference image dimension).

This is a convenience member function, equivalent to EnableLargeTranslations( !disable )

Warning
When large translations are enabled, the engine needs to double the sizes of internal working matrices, which quadruplicates the amount of required memory.
See also
EnableLargeTranslations(), AreLargeTranslationsEnabled()

Definition at line 372 of file FFTRegistration.h.

◆ EnableLargeTranslations()

void pcl::FFTTranslation::EnableLargeTranslations ( bool  enable = true)
inline

Enables or disables evaluation of large translations (>= one half of the reference image dimension).

Warning
When large translations are enabled, the engine needs to double the sizes of internal working matrices, which quadruplicates the amount of required memory.
See also
DisableLargeTranslations(), AreLargeTranslationsEnabled()

Definition at line 350 of file FFTRegistration.h.

◆ Peak()

float pcl::FFTTranslation::Peak ( ) const
inline

Returns the peak value read from the phase matrix for the last evaluated target image. The peak value isn't normalized, that is, it doesn't necessarily have to be in the [0,1] range.

Definition at line 418 of file FFTRegistration.h.


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