The Histogram class is a class for creating, managing, loading, and saving histogram for an Image. More...

#include <histogram.hpp>

Public Member Functions
	Histogram ()
	Histogram is the basic constructor setting variables to defaults. More...

	Histogram (Image *imgIn, VALUE_SPACE type, int nBin, int channel=0)
	Histogram is an extension of the basic constructor, where calculate is called in order to populate the Histogram. More...

	~Histogram ()
	~Histogram is the basic destructor which frees memory. More...

void	release ()
	release More...

void	calculate (Image imgIn, VALUE_SPACE type=VS_LIN, int nBin=256, BBox box=NULL, int channel=0)
	calculate computes the histogram of an input image. In the case of LDR images, they are ssumed to be normalized; i.e. with values in [0, 1]. This function computes the histogram for a single color channel. More...

void	uniform (float fMin, float fMax, uint value, VALUE_SPACE type, int nBin)
	uniform More...

void	update (float fMin, float fMax)
	update More...

int	project (float x)
	project converts an input value in the histogram domain. More...

float	unproject (int ind)
	unproject converts a histogram value back to its original domain. More...

void	ceiling (float k)
	ceiling limits the maximum value of the histogram using Ward algorithm. More...

void	clip (uint value)
	clip clips the histogram to value. More...

float *	cumulativef (bool bNormalized)
	cumulativef computes the cumulative Histogram. More...

float *	getCumulativef ()
	getCumulativef this function returns the cumulative Histogram. Note that cumulativef needs to be computed before otherwise the function will return a NULL pointer. More...

float	getfMin ()
	getfMin More...

float	getfMax ()
	getfMax More...

float *	getNormalized (bool bNor=true)
	getNormalized normalizes the Histogram. More...

float	getOtsu ()
	getOtsu More...

void	write (std::string name, bool bNor)
	write saves the Histogram as an Image into a file. More...

std::vector< float >	exposureCovering (int nBits=8, float overlap=1.0f)
	exposureCovering computes the exposure values for fully covering the dynamic range of the image. This function works only if the histogram was compute usign VS_LOG_2. More...

float	getBestExposure (int nBits, float overlap=0.5f)
	getBestExposure computes the best interval center. More...

Public Attributes
uint *	bin

uint *	bin_work

Protected Member Functions
float	projectDomain (float x)
	projectDomain applies the histogram domain to x. More...

float	unprojectDomain (float x)
	unprojectDomain removes the histogram domain to x. More...

void	update (float x)
	update More...

Protected Attributes
float *	bin_c

float *	bin_nor

int	nBin

VALUE_SPACE	type

float	fMin

float	fMax

float	deltaMaxMin

float	nBinf

float	epsilon

Detailed Description

The Histogram class is a class for creating, managing, loading, and saving histogram for an Image.

Constructor & Destructor Documentation

◆ Histogram() [1/2]

pic::Histogram::Histogram ( )

inline

Histogram is the basic constructor setting variables to defaults.

◆ Histogram() [2/2]

pic::Histogram::Histogram	(	Image *	imgIn,
		VALUE_SPACE	type,
		int	nBin,
		int	channel = `0`
	)

inline

Histogram is an extension of the basic constructor, where calculate is called in order to populate the Histogram.

Parameters

imgIn	is an input image for which Histogram needs to be computed.
type	is the space of computations (please see calculate()).
nBin	is the number of bins of the Histogram.
channel	is the color channel for which Histogram needs to be computed.

◆ ~Histogram()

pic::Histogram::~Histogram ( )

inline

~Histogram is the basic destructor which frees memory.

Member Function Documentation

◆ calculate()

void pic::Histogram::calculate	(	Image *	imgIn,
		VALUE_SPACE	type = `VS_LIN`,
		int	nBin = `256`,
		BBox *	box = `NULL`,
		int	channel = `0`
	)

inline

calculate computes the histogram of an input image. In the case of LDR images, they are ssumed to be normalized; i.e. with values in [0, 1]. This function computes the histogram for a single color channel.

Parameters

imgIn	is the input image for which the histogram needs to be computed
type	is the domain space for histogram computations. Histogram can be computed as: VS_LDR (256 bins), VS_LIN (linear space), VS_LOG_2 (logarithm 2 base), VS_LOG_E (natural logarithm space), and VS_LOG_10 (logarithm 10 base).
nBin	is the number of bins of the Histogram to be computed. The default value is 256.
box	is the slice where to compute the histogram.
channel	is the color channel for which the Histogram will be computed.

◆ ceiling()

void pic::Histogram::ceiling ( float k )

inline

ceiling limits the maximum value of the histogram using Ward algorithm.

Parameters

k

◆ clip()

void pic::Histogram::clip ( uint value )

inline

clip clips the histogram to value.

Parameters

value the maximum allowed value in the histogram.

◆ cumulativef()

float* pic::Histogram::cumulativef ( bool bNormalized )

inline

cumulativef computes the cumulative Histogram.

Parameters

bNormalized is a boolean value; if it is true values of the Histogram will be normalized.

Returns: It returns the cumulative Histogram as a float pointer.

◆ exposureCovering()

std::vector< float > pic::Histogram::exposureCovering	(	int	nBits = `8`,
		float	overlap = `1.0f`
	)

inline

exposureCovering computes the exposure values for fully covering the dynamic range of the image. This function works only if the histogram was compute usign VS_LOG_2.

Parameters

nBits	is the number of bit used for storing each output exposure image. The default value is 8.
overlap	is the value, in f-stops, of overlapping between two exposure images. This value is set to 1 by default.

Returns: It returns an std::vector<float> which contains the exposure values in f-stops for all required exposures for covering information.

◆ getBestExposure()

float pic::Histogram::getBestExposure	(	int	nBits,
		float	overlap = `0.5f`
	)

inline

getBestExposure computes the best interval center.

Parameters

nBits is the number of bits in the budget for the output image.

Returns: It returns the exposure, in f-stops, for setting the image with the best exposure at given dynamic range.

◆ getCumulativef()

float* pic::Histogram::getCumulativef ( )

inline

getCumulativef this function returns the cumulative Histogram. Note that cumulativef needs to be computed before otherwise the function will return a NULL pointer.

Returns: It returns a float pointer to the cumulative Histogram.

◆ getfMax()

float pic::Histogram::getfMax ( )

inline

getfMax

Returns

◆ getfMin()

float pic::Histogram::getfMin ( )

inline

getfMin

Returns

◆ getNormalized()

float* pic::Histogram::getNormalized ( bool bNor = true )

inline

getNormalized normalizes the Histogram.

Returns: It returns the normalized Histogram as a float pointer.

◆ getOtsu()

float pic::Histogram::getOtsu ( )

inline

getOtsu

Returns

◆ project()

int pic::Histogram::project ( float x )

inline

project converts an input value in the histogram domain.

Parameters

x	is an input value.

Returns: x is projected in the histogram domain.

◆ projectDomain()

float pic::Histogram::projectDomain ( float x )

inlineprotected

projectDomain applies the histogram domain to x.

Parameters

x	is an input value.

Returns: x is converted into the histogram domain.

◆ release()

void pic::Histogram::release ( )

inline

release

◆ uniform()

void pic::Histogram::uniform	(	float	fMin,
		float	fMax,
		uint	value,
		VALUE_SPACE	type,
		int	nBin
	)

inline

uniform

Parameters

value
type
nBin

◆ unproject()

float pic::Histogram::unproject ( int ind )

inline

unproject converts a histogram value back to its original domain.

Parameters

ind	is a histogram value.

Returns: ind is converted back to its original domain.

◆ unprojectDomain()

float pic::Histogram::unprojectDomain ( float x )

inlineprotected

unprojectDomain removes the histogram domain to x.

Parameters

x	is an input value.

Returns: x is converted back to its original domain.

◆ update() [1/2]

void pic::Histogram::update ( float x )

inlineprotected

update

Parameters

x

◆ update() [2/2]

void pic::Histogram::update	(	float	fMin,
		float	fMax
	)

inline

update

Parameters

fMin
fMax

◆ write()

void pic::Histogram::write	(	std::string	name,
		bool	bNor
	)

inline

write saves the Histogram as an Image into a file.

Parameters

name	is the filename where to save the Histogram.
bNor	is a boolean value for normalizing or not the Histogram.

Member Data Documentation

◆ bin

uint* pic::Histogram::bin

◆ bin_c

float* pic::Histogram::bin_c

protected

◆ bin_nor

float* pic::Histogram::bin_nor

protected

◆ bin_work

uint * pic::Histogram::bin_work

◆ deltaMaxMin

float pic::Histogram::deltaMaxMin

protected

◆ epsilon

float pic::Histogram::epsilon

protected

◆ fMax

float pic::Histogram::fMax

protected

◆ fMin

float pic::Histogram::fMin

protected

◆ nBin

int pic::Histogram::nBin

protected

◆ nBinf

float pic::Histogram::nBinf

protected

◆ type

VALUE_SPACE pic::Histogram::type

protected

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

C:/Users/Francesco/Documents/GitHub/piccante/include/histogram.hpp

Public Member Functions

Public Attributes

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ Histogram() [1/2]

◆ Histogram() [2/2]

◆ ~Histogram()

Member Function Documentation

◆ calculate()

◆ ceiling()

◆ clip()

◆ cumulativef()

◆ exposureCovering()

◆ getBestExposure()

◆ getCumulativef()

◆ getfMax()

◆ getfMin()

◆ getNormalized()

◆ getOtsu()

◆ project()

◆ projectDomain()

◆ release()

◆ uniform()

◆ unproject()

◆ unprojectDomain()

◆ update() [1/2]

◆ update() [2/2]

◆ write()

Member Data Documentation

◆ bin

◆ bin_c

◆ bin_nor

◆ bin_work

◆ deltaMaxMin

◆ epsilon

◆ fMax

◆ fMin

◆ nBin

◆ nBinf

◆ type