Displays various statistics from an image. More...

#include <ImageStatistics.h>

Inheritance diagram for casacore::ImageStatistics< T >:

Public Member Functions
	ImageStatistics (const ImageInterface< T > &image, LogIO &os, Bool showProgress=True, Bool forceDisk=False, Bool clone=True)
	Constructor takes the image and a `LogIO` object for logging. More...

	ImageStatistics (const ImageInterface< T > &image, Bool showProgress=True, Bool forceDisk=False, Bool clone=True)
	Constructor takes the image only. More...

	ImageStatistics (const ImageStatistics< T > &other)
	Copy constructor. More...

virtual	~ImageStatistics ()
	Destructor. More...

ImageStatistics< T > &	operator= (const ImageStatistics< T > &other)
	Assignment operator. More...

Bool	setNewImage (const ImageInterface< T > &image, Bool clone=True)
	Set a new ImageInterface object. More...

void	setPrecision (Int precision)

void	setBlc (const IPosition &blc)

IPosition	getBlc () const

Int	getPrecision () const

void	showRobust (const Bool show)
	list robust statistics? Should be called before display() More...

void	recordMessages (const Bool rm)

vector< String >	getMessages ()

void	clearMessages ()

void	setListStats (Bool b)

Public Member Functions inherited from casacore::LatticeStatistics< T >
	LatticeStatistics (const MaskedLattice< T > &lattice, LogIO &os, Bool showProgress=True, Bool forceDisk=False, Bool clone=True)
	Constructor takes the lattice and a `LogIO` object for logging. More...

	LatticeStatistics (const MaskedLattice< T > &lattice, Bool showProgress=True, Bool forceDisk=False, Bool clone=True)
	Constructor takes the lattice only. More...

	LatticeStatistics (const LatticeStatistics< T > &other)
	Copy constructor. More...

virtual	~LatticeStatistics ()
	Destructor. More...

LatticeStatistics< T > &	operator= (const LatticeStatistics< T > &other)
	Assignment operator. More...

Bool	setAxes (const Vector< Int > &cursorAxes)
	Set the cursor axes (0 relative). More...

Bool	setInExCludeRange (const Vector< T > &include, const Vector< T > &exclude, Bool setMinMaxToInclude=False)
	You may specify a pixel intensity range as either one for which all pixels in that range are included or one for which all pixels in that range are excluded. More...

Bool	setList (const Bool &doList)
	This function allows you to control whether the statistics are written to the output stream if you are also making a plot. More...

Bool	display ()
	Display the statistics by listing and/or plotting them. More...

Bool	getLayerStats (String &stats, Double area, Int zAxis=-1, Int zLayer=-1, Int hAxis=-1, Int hLayer=-1)

Bool	getLayerStats (stat_list &stats, Double area, Int zAxis=-1, Int zLayer=-1, Int hAxis=-1, Int hLayer=-1)

Vector< Int >	displayAxes () const
	Return the display axes. More...

Bool	getStatistic (Array< AccumType > &stat, LatticeStatsBase::StatisticsTypes type, Bool dropDeg=True)
	Recover the desired Statistic into an array. More...

Bool	getConvertedStatistic (Array< T > &stat, LatticeStatsBase::StatisticsTypes type, Bool dropDeg=True)

Bool	getMinMaxPos (IPosition &minPos, IPosition &maxPos)
	Recover position of min and max. More...

Bool	getStats (Vector< AccumType > &, const IPosition &pos, const Bool posInLattice=False)
	This function gets a vector containing all the statistics for a given location. More...

void	resetError ()
	Reset argument error condition. More...

Bool	getFullMinMax (T &dataMin, T &dataMax)
	Get full lattice min and max only. More...

String	errorMessage () const
	Recover last error message. More...

Bool	setNewLattice (const MaskedLattice< T > &lattice, Bool clone=True)
	Set a new MaskedLattice object. More...

Bool	hasLogger () const
	Did we construct with a logger ? More...

Bool	configureBiweight (Int maxIter, Double c)
	The configure methods return True if reconfiguration is actually necessary (ie if the underlying storage lattice needs to be recomputed). More...

Bool	configureClassical ()
	configure object to use Classical Statistics The time, t_x, it takes to compute classical statistics using algorithm x, can be modeled by t_x = n_sets(a_x + b_xn_el) where n_sets is the number of independent sets of data to compute stats on, each containing n_el number of elements. More...

Bool	configureClassical (Double aOld, Double bOld, Double aNew, Double bNew)

Bool	configureFitToHalf (FitToHalfStatisticsData::CENTER centerType=FitToHalfStatisticsData::CMEAN, FitToHalfStatisticsData::USE_DATA useData=FitToHalfStatisticsData::LE_CENTER, AccumType centerValue=0)
	configure to use fit to half algorithm. More...

Bool	configureHingesFences (Double f)
	configure to use hinges-fences algorithm More...

Bool	configureChauvenet (Double zscore=-1, Int maxIterations=-1)
	configure to use Chauvenet's criterion More...

void	forceUseStatsFrameworkUsingDataProviders ()
	The force* methods are really only for testing. More...

void	forceUseStatsFrameworkUsingArrays ()

void	forceUseOldTiledApplyMethod ()

void	forceAllowCodeDecideWhichAlgortihmToUse ()

std::map< String, uInt >	getChauvenetNiter () const
	get number of iterations associated with Chauvenet criterion algorithm More...

void	setComputeQuantiles (Bool b)
	should quantile-like stats (median, quartiles, medabsdevmed) be computed? When the stats framework is used, It is better to set this before computing any statistics, to avoid unnecessary duplicate creations of the stats algorithm objects. More...

Protected Types
typedef NumericTraits< T > ::PrecisionType	AccumType

Protected Member Functions
virtual Bool	_canDoFlux () const
	Virtual Functions. More...

Protected Member Functions inherited from casacore::LatticeStatistics< T >
virtual Bool	listLayerStats (const Matrix< AccumType > &ord, ostringstream &rslt, Int zLayer)

IPosition	locInLattice (const IPosition &storagePosition, Bool relativeToParent=True) const
	Given a location in the storage lattice, convert those locations on the non-statistics axis (the last one) and optionally account for the lattice subsectioning. More...

void	setStream (ostream &os, Int oPrec)
	Non-virtual functions. More...

IPosition	_storageLatticeShape () const
	get the storage lattice shape More...

void	_latticePosToStoragePos (IPosition &storagePos, const IPosition &latticePos)
	convert a position in the input lattice to the corresponding position in the stats storage lattice. More...

StatisticsData::ALGORITHM	_getAlgorithm () const

Private Member Functions
virtual void	getLabels (String &higherOrder, String &xAxis, const IPosition &dPos) const
	Virtual functions. More...

virtual Bool	_getBeamArea (Array< Double > &beamArea, String &msg) const
	Get beam area in pixels if possible. More...

virtual void	listMinMax (ostringstream &osMin, ostringstream &osMax, Int oWidth, DataType type)
	List min and max with world coordinates. More...

virtual Bool	listStats (Bool hasBeam, const IPosition &dPos, const Matrix< AccumType > &ord)
	List the statistics. More...

virtual void	displayStats (AccumType nPts, AccumType sum, AccumType median, AccumType medAbsDevMed, AccumType quartile, AccumType sumSq, AccumType mean, AccumType var, AccumType rms, AccumType sigma, AccumType dMin, AccumType dMax, AccumType q1, AccumType q3)

Quantum< AccumType >	_flux (Bool &isFluxDensity, AccumType sum, Double beamAreaInPixels) const
	If `isFluxDensity` is False, then the computed value is a flux (ie flux density integrated over a spectral extent) More...

Bool	_computeFlux (Array< AccumType > &flux, const Array< AccumType > &npts, const Array< AccumType > &sum)

Bool	_computeFlux (Quantum< AccumType > &flux, AccumType sum, const IPosition &pos, Bool posInLattice)

Private Attributes
LogIO	os_p
	Data. More...

const ImageInterface< T > *	pInImage_p

std::shared_ptr< const ImageInterface< T > >	_inImPtrMgr

IPosition	blc_

Int	precision_

Bool	_showRobust

Bool	_recordMessages

Bool	_listStats

vector< String >	_messages

Additional Inherited Members
Public Types inherited from casacore::LatticeStatistics< T >
typedef NumericTraits< T > ::PrecisionType	AccumType

typedef std::pair< String, String >	stat_element

typedef std::list< stat_element >	stat_list

Public Types inherited from casacore::LatticeStatsBase
enum	StatisticsTypes { NPTS, SUM, SUMSQ, MEDIAN, MEDABSDEVMED, QUARTILE, Q1, Q3, MIN, MAX, MEAN, VARIANCE, SIGMA, RMS, FLUX, NSTATS, NACCUM }
	This `enum StatisticTypes` is provided for use with the `LatticeStatistics<T>::setPlotting` function. More...

Static Public Member Functions inherited from casacore::LatticeStatsBase
static Vector< Int >	toStatisticTypes (const String &statistics, const Regex &delimiter)
	Helper function to convert a String containing a list of desired statistics to the correct Vector<Int> required for the LatticeStatistics<T>::setPlotting function. More...

static Vector< Int >	toStatisticTypes (const Vector< String > &statistics)

static String	toStatisticName (StatisticsTypes type)
	Convert type to string. More...

static String	toStatisticName (Int type)

static Int	toStatisticType (const String &statistic)
	Returns -1 if the statistic string is not valid. More...

static Bool	setNxy (Vector< Int > &nxy, ostream &os)
	Check and fill in defaults for a `Vector<Int>` containing the number of subplots in x and y to be put on a plot. More...

static void	setStorageImageShape (IPosition &storeImageShape, const Bool &last, const Int &axisSize, const Vector< Int > &displayAxes, const IPosition &shape)
	A storage image is used to accumulate information as a function of the display axes as an image is iterated through. More...

static void	stretchMinMax (Float &min, Float &max)
	Stretch a range by 10%. More...

static std::set< Double >	quartileFracs ()

Protected Attributes inherited from casacore::LatticeStatistics< T >
LogIO	os_p

Vector< Int >	cursorAxes_p

Vector< Int >	displayAxes_p

Bool	goodParameterStatus_p

Bool	haveLogger_p

Bool	fixedMinMax_p

Bool	doRobust_p
	doRobust means that when the storage lattice is generated, the robust statistics are generated as well More...

Bool	doList_p

IPosition	minPos_p

IPosition	maxPos_p

IPosition	blcParent_p

String	error_p

Detailed Description

template<class T>
class casacore::ImageStatistics< T >

Displays various statistics from an image.

Intended use:

Public interface

Review Status

Date Reviewed:: yyyy/mm/dd

Prerequisite

LatticeStatistics (base class)
ImageInterface

Etymology

This is a class designed to display and retrieve statistics from images

Synopsis

This class enable you to display and/or retrieve statistics evaluated over specified regions from an image. The dimension of the region is arbitrary, but the size of each dimension is always the size of the corresponding image axis. The statistics are displayed as a function of location of the axes not used to evaluate the statistics over. The axes which you evaluate the statistics over are called the cursor axes, the others are called the display axes.

This class is derived from the class LatticeStatistics which does all the work. This class only adds some extra capability in terms of logging world (rather than pixel) coordinates and computing the synthesized beam area, if there is one. There are just a few virtual functions for you to over-ride. These are rather specialized, they are not part of a general polymorphic interface, just a way to separate the Lattice and Image functionality out.

See LatticeStatistics for details and examples.

Motivation

The generation of statistical information from an image is a basic and necessary capability.

To Do

Deal with complex images at least for statistics retrieval if not plotting.
Retrieve statistics at specified location of display axes
Standard errors on statistical quantities
Median, other more exotic statistics. Life made difficult by accumulation image approach

Definition at line 99 of file ImageStatistics.h.

Member Typedef Documentation

template<class T>

typedef NumericTraits<T>::PrecisionType casacore::ImageStatistics< T >::AccumType

protected

Definition at line 173 of file ImageStatistics.h.

Constructor & Destructor Documentation

template<class T>

casacore::ImageStatistics< T >::ImageStatistics	(	const ImageInterface< T > &	image,
		LogIO &	os,
		Bool	showProgress = `True`,
		Bool	forceDisk = `False`,
		Bool	clone = `True`
	)

Constructor takes the image and a LogIO object for logging.

You can specify whether you want to see progress meters or not. You can force the storage image to be disk based, otherwise the decision for core or disk is taken for you. If clone is True, the input image will be cloned, so the caller can make changes to the input image, but the statistics will reflect the image as it was at construction. If False, a reference to the input image is used, and so the caller shouldn't make changes to the input image between construction and calling statistics computation methods, unless it calls setNewImage() to update the changed image. Obviously, cloning the image impacts performance and memory usage.

template<class T>

casacore::ImageStatistics< T >::ImageStatistics	(	const ImageInterface< T > &	image,
		Bool	showProgress = `True`,
		Bool	forceDisk = `False`,
		Bool	clone = `True`
	)

Constructor takes the image only.

In the absence of a logger you get no messages. This includes error messages and potential listing of the statistics. You can specify whether you want to see progress meters or not. You can force the storage image to be disk based, otherwise the decision for core or disk is taken for you.

template<class T>

casacore::ImageStatistics< T >::ImageStatistics ( const ImageStatistics< T > & other )

Copy constructor.

Copy semantics are followed. Therefore any storage image that has already been created for other is copied to *this

template<class T>

virtual casacore::ImageStatistics< T >::~ImageStatistics ( )

virtual

Destructor.

Member Function Documentation

template<class T>

virtual Bool casacore::ImageStatistics< T >::_canDoFlux ( ) const

protectedvirtual

Virtual Functions.

See implementation to figure it all out !

FIXME The indirect dependence of this class on ImageInterface related issues (eg flux density) breaks encapsulation. All the ImageInterface related code should be encapsulated in ImageStatistics. Unfortunately, that requires significantly more time than I have atm. A return value of False means that the object in question cannot compute flux density values. The default implementation returns False.

Reimplemented from casacore::LatticeStatistics< T >.

template<class T>

Bool casacore::ImageStatistics< T >::_computeFlux	(	Array< AccumType > &	flux,
		const Array< AccumType > &	npts,
		const Array< AccumType > &	sum
	)

privatevirtual

Reimplemented from casacore::LatticeStatistics< T >.

template<class T>

Bool casacore::ImageStatistics< T >::_computeFlux	(	Quantum< AccumType > &	flux,
		AccumType	sum,
		const IPosition &	pos,
		Bool	posInLattice
	)

privatevirtual

Reimplemented from casacore::LatticeStatistics< T >.

template<class T>

Quantum<AccumType> casacore::ImageStatistics< T >::_flux	(	Bool &	isFluxDensity,
		AccumType	sum,
		Double	beamAreaInPixels
	)		const

privatevirtual

If isFluxDensity is False, then the computed value is a flux (ie flux density integrated over a spectral extent)

Reimplemented from casacore::LatticeStatistics< T >.

template<class T>

virtual Bool casacore::ImageStatistics< T >::_getBeamArea	(	Array< Double > &	beamArea,
		String &	msg
	)		const

privatevirtual

Get beam area in pixels if possible.

Return False if the beam area could not be calculated.

template<class T>

void casacore::ImageStatistics< T >::clearMessages ( )

inline

Definition at line 168 of file ImageStatistics.h.

References casacore::ImageStatistics< T >::_messages.

template<class T>

virtual void casacore::ImageStatistics< T >::displayStats	(	AccumType	nPts,
		AccumType	sum,
		AccumType	median,
		AccumType	medAbsDevMed,
		AccumType	quartile,
		AccumType	sumSq,
		AccumType	mean,
		AccumType	var,
		AccumType	rms,
		AccumType	sigma,
		AccumType	dMin,
		AccumType	dMax,
		AccumType	q1,
		AccumType	q3
	)

privatevirtual

Reimplemented from casacore::LatticeStatistics< T >.

template<class T>

IPosition casacore::ImageStatistics< T >::getBlc ( ) const

template<class T>

virtual void casacore::ImageStatistics< T >::getLabels	(	String &	higherOrder,
		String &	xAxis,
		const IPosition &	dPos
	)		const

privatevirtual

Virtual functions.

See LatticeStatistics for more information about these, or see the implementation.

Get label for higher order axes

template<class T>

vector<String> casacore::ImageStatistics< T >::getMessages ( )

inline

Definition at line 166 of file ImageStatistics.h.

References casacore::ImageStatistics< T >::_messages.

template<class T>

Int casacore::ImageStatistics< T >::getPrecision ( ) const

template<class T>

virtual void casacore::ImageStatistics< T >::listMinMax	(	ostringstream &	osMin,
		ostringstream &	osMax,
		Int	oWidth,
		DataType	type
	)

privatevirtual

List min and max with world coordinates.

Reimplemented from casacore::LatticeStatistics< T >.

template<class T>

virtual Bool casacore::ImageStatistics< T >::listStats	(	Bool	hasBeam,
		const IPosition &	dPos,
		const Matrix< AccumType > &	ord
	)

privatevirtual

List the statistics.

Reimplemented from casacore::LatticeStatistics< T >.

template<class T>

ImageStatistics<T>& casacore::ImageStatistics< T >::operator= ( const ImageStatistics< T > & other )

Assignment operator.

Deletes any storage image associated with the object being assigned to and copies any storage image that has already been created for "other".

template<class T>

void casacore::ImageStatistics< T >::recordMessages ( const Bool rm )

inline

Definition at line 164 of file ImageStatistics.h.

References casacore::ImageStatistics< T >::_recordMessages.

template<class T>

void casacore::ImageStatistics< T >::setBlc ( const IPosition & blc )

template<class T>

void casacore::ImageStatistics< T >::setListStats ( Bool b )

inline

Definition at line 170 of file ImageStatistics.h.

References casacore::ImageStatistics< T >::_listStats.

template<class T>

Bool casacore::ImageStatistics< T >::setNewImage	(	const ImageInterface< T > &	image,
		Bool	clone = `True`
	)

Set a new ImageInterface object.

A return value of False indicates the image had an invalid type or that the internal state of the class is bad. If clone is True, the input image will be cloned, so the caller can make changes to the input image, but the statistics will reflect the image as it was at construction. If False, a reference to the input image is used, and so the caller shouldn't make changes to the input image between construction and calling statistics computation methods, unless it calls setNewImage() to update the changed image. Obviously, cloning the image impacts performance and memory usage.

template<class T>

void casacore::ImageStatistics< T >::setPrecision ( Int precision )