The GeometryDecoder class provides the functionality to extract the system configuration from a set of projection matrices. More...

#include <geometrydecoder.h>

Public Types
enum	PhysicalDimension { PixelWidth, PixelHeight, SourceDetectorDistance }

Public Member Functions
	GeometryDecoder (const QSize &pixelPerModule, const QSizeF &pixelSize)

	GeometryDecoder (const QSize &pixelPerModule, PhysicalDimension physicalDimension=PixelWidth, double mm=1.0)

AcquisitionSetup	decodeFullGeometry (const FullGeometry &geometry) const

const QPair< PhysicalDimension, double > &	dimensionReference () const

const QSize &	pixelPerModule () const

const QSizeF &	pixelDimensions () const

const QSizeF &	pixelSize () const

void	setDimensionReference (PhysicalDimension physicalDimension, double mm)

void	setPixelPerModule (const QSize &value)

void	setPixelDimensions (const QSizeF &value)

void	setPixelSize (const QSizeF &value)

void	setReferenceModuleID (int moduleID)

Static Public Member Functions
static SimpleCTSystem	decodeSingleViewGeometry (const SingleViewGeometry &singleViewGeometry, const QSize &pixelPerModule, PhysicalDimension physicalDimension=PixelWidth, double mm=1.0, int referenceModuleID=-1)

static AcquisitionSetup	decodeFullGeometry (const FullGeometry &geometry, const QSize &pixelPerModule, const QSizeF &pixelSize, int referenceModuleID=-1)

static AcquisitionSetup	decodeFullGeometry (const FullGeometry &geometry, const QSize &pixelPerModule, PhysicalDimension physicalDimension=PixelWidth, double mm=1.0, int referenceModuleID=-1)

Static Private Member Functions
static QSizeF	computePixelSize (const mat::ProjectionMatrix &pMat, PhysicalDimension physicalDimension, double mm)

static QVector< mat::Location >	computeModuleLocations (const SingleViewGeometry &singleViewGeometry, const Vector3x1 &sourcePosition, const QSize &pixelPerModule, double pixelWidth)

static Matrix3x3	computeSourceRotation (const QVector< mat::Location > &moduleLocations, const Vector3x1 &sourcePosition)

static mat::Location	referenceModuleLocation (const QVector< mat::Location > &moduleLocations, int referenceModuleID)

static void	adaptModuleLocationsToReference (QVector< mat::Location > &moduleLocations, const mat::Location &referenceLocation)

Private Attributes
QSize	_pixelPerModule
	Number of pixels of the assumed detector.

QSizeF	_pixelSize
	Pixel size of the assumed detector.

QPair< PhysicalDimension, double >	_physicalDimensionReference = { PixelWidth, 1.0 }

int	_referenceModule = -1

Detailed Description

The GeometryDecoder class provides the functionality to extract the system configuration from a set of projection matrices.

This class can be used to generate an AcquisitionSetup that fully represents a set of given projection matrices. Besides the projection matrices themselves, information about the detector dimensions (i.e. number of pixels in each module and their (physical) size) is required.

The projection matrices must be passed as a FullGeometry object. In doing so, the routine can automatically readout the number of individual modules in the detector.

Constructor & Destructor Documentation

◆ GeometryDecoder()

CTL::GeometryDecoder::GeometryDecoder	(	const QSize &	pixelPerModule,
		const QSizeF &	pixelSize
	)

Constructs a GeometryDecoder object.

Member Function Documentation

◆ computeModuleLocations()

QVector< mat::Location > CTL::GeometryDecoder::computeModuleLocations	(	const SingleViewGeometry &	singleViewGeometry,
		const Vector3x1 &	sourcePosition,
		const QSize &	pixelPerModule,
		double	pixelWidth
	)

staticprivate

Computes the physical center of the modules in world coordinates.

◆ computeSourceRotation()

Matrix3x3 CTL::GeometryDecoder::computeSourceRotation	(	const QVector< mat::Location > &	moduleLocations,
		const Vector3x1 &	sourcePosition
	)

staticprivate

Returns matrix that rotates source perpendicular to the centroid of detector locations

◆ decodeFullGeometry() [1/2]

AcquisitionSetup CTL::GeometryDecoder::decodeFullGeometry ( const FullGeometry & geometry ) const

See decodeFullGeometry(const FullGeometry& geometry, const QSize& pixelPerModule, const QSizeF& pixelSize). It uses the internal _pixelPerModule and _pixelSize of the object.

◆ decodeFullGeometry() [2/2]

AcquisitionSetup CTL::GeometryDecoder::decodeFullGeometry	(	const FullGeometry &	geometry,
		const QSize &	pixelPerModule,
		const QSizeF &	pixelSize,
		int	referenceModuleID = `-1`
	)

static

Decodes the set of projection matrices in geometry and constructs a GenericAcquisitionSetup that represents all the geometry information that has been extracted.

Besides projection matrices, information about the detector dimensions (i.e. number of pixels in each module and their (physical) size) is required. The number of individual flat panel modules is automatically readout from geometry.

This method constructs a SimpleCTSystem consisting of a GenericSource, GenericDetector and a GenericGantry component.

Some remarks on system configuration: All source settings remain at default values (i.e. source spectrum, focal spot size and focal spot position). Consider changing these afterwards, if required. The full geometry information regarding the detector is stored in the location specification of the individual detector modules. In particular, this means that the (global) detector positioning - as queried for example by GenericGantry::detectorPosition() - will carry no information (position defaults to (0,0,0) and rotation to identity matrix). Additionally, the rotation of the source component cannot be determined without further information. Hence, it remains at default value (i.e. identity matrix).

◆ pixelDimensions()

const QSizeF & CTL::GeometryDecoder::pixelDimensions ( ) const

This function has been deprecated. Use pixelSize() instead.

◆ pixelPerModule()

const QSize & CTL::GeometryDecoder::pixelPerModule ( ) const

Returns the number of pixels per module of the detector that the geometry decoder shall assume for the system.

◆ pixelSize()

const QSizeF & CTL::GeometryDecoder::pixelSize ( ) const

Returns the pixel size [mm] of the detector that the geometry decoder shall assume for the system.

◆ setPixelDimensions()

void CTL::GeometryDecoder::setPixelDimensions ( const QSizeF & value )

This function has been deprecated. Use setPixelSize(const QSizeF& value) instead.

◆ setPixelPerModule()

void CTL::GeometryDecoder::setPixelPerModule ( const QSize & value )

Sets the number of pixels per module of the detector that the geometry decoder shall assume for the system to value.

◆ setPixelSize()

void CTL::GeometryDecoder::setPixelSize ( const QSizeF & value )

Sets the pixel size [mm] of the detector that the geometry decoder shall assume for the system to value.

◆ setReferenceModuleID()

void CTL::GeometryDecoder::setReferenceModuleID ( int moduleID )

Sets the reference module that the geometry decoder shall use to compute the detector location to moduleID. If moduleID < 0, automatic selection will be performed, i.e. the central module ID (nbModules / 2) will be used.

Note that moduleID must be a valid id in the data to be decoded (i.e. moduleID < nbModules).

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

modules/src/acquisition/geometrydecoder.h
modules/src/acquisition/geometrydecoder.cpp

Public Types

Public Member Functions

Static Public Member Functions

Static Private Member Functions

Private Attributes