GaussianNoiseModel Class Reference

Gaussian noise class. More...

#include <GaussianNoiseModel.hh>

Inherits Noise.

Inherited by ImageGaussianNoiseModel.

Public Types

enum  NoiseType { NONE, CUSTOM, GAUSSIAN }
 Which noise types we support. More...
 

Public Member Functions

 GaussianNoiseModel ()
 Constructor. More...
 
virtual ~GaussianNoiseModel ()
 Destructor. More...
 
double Apply (double _in)
 Apply noise to input data value. More...
 
double ApplyImpl (double _in)
 Apply noise to input data value. More...
 
virtual void Fini ()
 Finalize the noise model. More...
 
double GetBias () const
 Accessor for bias. More...
 
double GetMean () const
 Accessor for mean. More...
 
NoiseType GetNoiseType () const
 Accessor for NoiseType. More...
 
double GetStdDev () const
 Accessor for stddev. More...
 
virtual void Load (sdf::ElementPtr _sdf)
 Load noise parameters from sdf. More...
 
virtual void Print (std::ostream &_out) const
 Documentation inherited. More...
 
virtual void SetCamera (rendering::CameraPtr _camera)
 Set camera needed to create image noise. More...
 
virtual void SetCustomNoiseCallback (boost::function< double(double)> _cb)
 Register a custom noise callback. More...
 

Protected Attributes

double bias
 If type starts with GAUSSIAN, the bias we'll add. More...
 
double mean
 If type starts with GAUSSIAN, the mean of the distribution from which we sample when adding noise. More...
 
double precision
 If type==GAUSSIAN_QUANTIZED, the precision to which the output signal is rounded. More...
 
bool quantized
 True if the type is GAUSSIAN_QUANTIZED. More...
 
double stdDev
 If type starts with GAUSSIAN, the standard deviation of the distribution from which we sample when adding noise. More...
 

Detailed Description

Gaussian noise class.

Gaussian noise class for image sensors.

Member Enumeration Documentation

enum NoiseType
inherited

Which noise types we support.

Enumerator
NONE 
CUSTOM 
GAUSSIAN 

Constructor & Destructor Documentation

Constructor.

virtual ~GaussianNoiseModel ( )
virtual

Destructor.

Member Function Documentation

double Apply ( double  _in)
inherited

Apply noise to input data value.

Parameters
[in]_inInput data value.
Returns
Data with noise applied.
double ApplyImpl ( double  _in)
virtual

Apply noise to input data value.

This gets overriden by derived classes, and called by Apply.

Parameters
[in]_inInput data value.
Returns
Data with noise applied.

Reimplemented from Noise.

virtual void Fini ( )
virtual

Finalize the noise model.

Reimplemented from Noise.

Reimplemented in ImageGaussianNoiseModel.

double GetBias ( ) const

Accessor for bias.

Returns
Bias on output.
double GetMean ( ) const

Accessor for mean.

Returns
Mean of Gaussian noise.
NoiseType GetNoiseType ( ) const
inherited

Accessor for NoiseType.

Returns
Type of noise currently in use.
double GetStdDev ( ) const

Accessor for stddev.

Returns
Standard deviation of Gaussian noise.
virtual void Load ( sdf::ElementPtr  _sdf)
virtual

Load noise parameters from sdf.

Parameters
[in]_sdfSDF parameters.
[in]_sensorType of sensor.

Reimplemented from Noise.

Reimplemented in ImageGaussianNoiseModel.

virtual void Print ( std::ostream &  _out) const
virtual

Documentation inherited.

Reimplemented from Noise.

Reimplemented in ImageGaussianNoiseModel.

virtual void SetCamera ( rendering::CameraPtr  _camera)
virtualinherited

Set camera needed to create image noise.

This is only needed for image sensors, i.e. camera/multicamera/depth sensors, which use shaders for more efficient noise generation.

Parameters
[in]_cameraCamera associated to an image sensor

Reimplemented in ImageGaussianNoiseModel.

virtual void SetCustomNoiseCallback ( boost::function< double(double)>  _cb)
virtualinherited

Register a custom noise callback.

Parameters
[in]_cbCallback function for applying a custom noise model. This is useful if users want to use their own noise model from a sensor plugin.

Member Data Documentation

double bias
protected

If type starts with GAUSSIAN, the bias we'll add.

double mean
protected

If type starts with GAUSSIAN, the mean of the distribution from which we sample when adding noise.

double precision
protected

If type==GAUSSIAN_QUANTIZED, the precision to which the output signal is rounded.

bool quantized
protected

True if the type is GAUSSIAN_QUANTIZED.

double stdDev
protected

If type starts with GAUSSIAN, the standard deviation of the distribution from which we sample when adding noise.


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